Asymmetric Organic Synthesis with Enzymes

Asymmetric Organic Synthesis with Enzymes Edited by Vicente Gotor, Ignacio Alfonso, and Eduardo Garcı´a-Urdiales Rel...

Author: Vicente Gotor | Ignacio Alfonso | Eduardo García-Urdiales

85 downloads 1581 Views 3MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!

Report copyright / DMCA form

DOWNLOAD PDF

Asymmetric Organic Synthesis with Enzymes

Edited by Vicente Gotor, Ignacio Alfonso, and Eduardo Garcı´a-Urdiales

Related Titles Eames, J., Todd, M. (eds.)

Cao, L.

Separation of Enantiomers

Carrier-bound Immobilized Enzymes

New Methods and Applications approx. 400 pages Hardcover ISBN: 978-3-527-30502-5

Aehle, W. (ed.)

Enzymes in Industry Production and Applications approx. 520 pages with approx. 120 ﬁgures 2007 Hardcover ISBN: 978-3-527-31689-2

Principles, Application and Design 578 pages with 248 ﬁgures and 67 tables 2005 Hardcover ISBN: 978-3-527-31232-0

Breslow, R. (ed.)

Artificial Enzymes 193 pages with 81 ﬁgures and 8 tables 2005 Hardcover ISBN: 978-3-527-31165-1

Reymond, J.-L. (ed.)

Enzyme Assays High-throughput Screening, Genetic Selection and Fingerprinting 386 pages with 162 ﬁgures and 12 tables 2006 Hardcover ISBN: 978-3-527-31095-1

Bisswanger, H.

Practical Enzymology 272 pages with 35 ﬁgures and 6 tables 2004 Softcover ISBN: 978-3-527-30444-8

Bommarius, A.S.,Riebel-Bommarius,B.R.

Bornscheuer, U.T., Kazlauskas, R.J.

Biocatalysis

Hydrolases in Organic Synthesis

Fundamentals and Applications

Regio- and Stereoselective Biotransformations

2004 Hardcover ISBN: 978-3-527-30344-1

368 pages with 255 ﬁgures and 23 tables 2006 Hardcover ISBN: 978-3-527-31029-6

634 pages with 251 ﬁgures and 78 tables

Asymmetric Organic Synthesis with Enzymes

Edited by Vicente Gotor, Ignacio Alfonso, and Eduardo García-Urdiales

The Editors Prof. Vicente Gotor Dept.de Química Orgánica/Inorg Universidad de Oviedo Avenida Julián Clavería 6 33006 Oviedo Spain Prof. Ignacio Alfonso Dept.de Química Orgánica/Inorg Universidad Jaume I Campus del Riu Sec 12071 Castellón Spain Dr. Eduardo García-Urdiales EMBL - Biocomputing Meyerhofstr. 1 69117 Heidelberg Germany

All books published by Wiley-VCH are carefully produced. Nevertheless, authors, editors, and publisher do not warrant the information contained in these books, including this book, to be free of errors. Readers are advised to keep in mind that statements, data, illustrations, procedural details or other items may inadvertently be inaccurate. Library of Congress Card No.: applied for British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library. Bibliographic information published by the Deutsche Nationalbibliothek Die Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliograﬁe; detailed bibliographic data are available in the Internet at . # 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim All rights reserved (including those of translation into other languages). No part of this book may be reproduced in any form – by photoprinting, microﬁlm, or any other means – nor transmitted or translated into a machine language without written permission from the publishers. Registered names, trademarks, etc. used in this book, even when not speciﬁcally marked as such, are not to be considered unprotected by law. Typesetting Thomson Digital, Noida, India Printing Strauss GmbH, Mörlenbach Binding Litges & Dopf GmbH, Heppenheim Printed in the Federal Republic of Germany Printed on acid-free paper ISBN: 978-3-527-31825-4

V

Contents Preface X1 List of Contributors

XIII

1

I

Methodology

1

Medium Engineering 3 Giacomo Carrea and Sergio Riva Introduction 3 Modulation of Enzyme Enantioselectivity by Medium Engineering 5 Selectivity Enhancement by Addition of Water-Miscible Organic Cosolvents 5 Selectivity Enhancement in Organic Media with Low Water Activity 7 Organic Solvent Systems 7 Enzyme Properties in Organic Solvents 8 Medium Engineering 9 Rationales 12 Modulation of Enzyme Selectivity: New Trends of Research 14 Ionic Liquids 14 Additives 16 Conclusions and Outlooks 17 References 17

1.1 1.2 1.2.1 1.2.2 1.2.2.1 1.2.2.2 1.2.2.3 1.2.3 1.2.4 1.2.4.1 1.2.4.2 1.3

2

2.1 2.2 2.3

Directed Evolution as a Means to Engineer Enantioselective Enzymes 21 Manfred T. Reetz Introduction 21 Molecular Biological Methods for Mutagenesis 23 High-throughput Screening Methods for Enantioselectivity

Asymmetric Organic Synthesis with Enzymes. Edited by Vicente Gotor, Ignacio Alfonso, and Eduardo García-Urdiales Copyright Ó 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 978-3-527-31825-4

27

VI

Contents

2.4 2.4.1 2.4.2 2.4.3 2.4.4 2.4.5 2.4.6 2.4.7 2.4.8 2.4.9 2.4.10 2.4.11 2.4.12 2.4.13 2.5

3 3.1 3.2 3.2.1 3.2.2 3.2.3 3.3 3.3.1 3.3.1.1 3.3.1.2 3.3.1.3 3.3.1.4 3.3.2 3.3.2.1 3.3.2.2 3.3.2.3 3.3.2.4 3.3.2.5 3.3.2.6 3.3.2.7 3.3.2.8 3.4

Examples of Enhancing the Enantioselectivity of Enzymes by Directed Evolution 28 Lipase from Pseudomonas aeruginosa (PAL) 28 Other Lipases 38 Esterases 38 Hydantoinases 39 Nitrilases 39 Epoxide Hydrolases 41 Phosphotriesterases 45 Aminotransferases 45 Aldolases 46 Cyclohexanone and Cyclopentanone Monooxygenases as Baeyer–Villigerases and Sulfoxidation Catalysts 48 Monoamine Oxidases 54 Cytochrome P450 Enzymes 55 Other Enzymes 55 Conclusions and Perspectives 56 References 56 The Search for New Enzymes 65 Jean-Louis Reymond and Wolfgang Streit Introduction 65 Mechanism-based Enzyme Design 66 Catalytic Antibodies 66 Rational Design of New Catalysts on Enzyme and Protein Basis Synthetic Enzyme Models 71 Metagenomics 71 Construction of Metagenome-derived DNA Libraries 72 Selection of the Environment 72 Cloning Strategies 73 Screening and Detection Technologies 73 Major Problems that Need to be Addressed 74 The Genomes of Not Yet Cultured Microbes as Resources for Novel Genes 75 Polysaccharide Degrading/Modifying Enzymes 75 Lipolytic Biocatalysts 77 Vitamin Biosynthesis 77 Nitrilases, Nitrile Hydratases, and Amidases 78 Oxidoreductases/Dehydrogenases 79 Proteases 79 Glycerol Hydratases 79 Antibiotics and Pharmaceuticals 79 Conclusion 80 References 80

69

Contents

87

II

Synthetic Applications

4

Dynamic Kinetic Resolutions 89 Belén Martín-Matute and Jan-E. Bäckvall Introduction 89 Synthesis of Enantiomerically Pure Compounds 89 Kinetic Resolution (KR) and Dynamic Kinetic Resolution (DKR) 90 Enzymes in Organic Chemistry 91 Metal-Catalyzed Racemization 92 DKR of Allylic Acetates and Allylic Alcohols 93 DKR of sec-alcohols 94 DKR of Amines 98 Base-Catalyzed Racemization 98 DKR of Thioesters 99 DKR of Activated Esters 100 DKR of Oxazolones 100 DKR of Hydantoins 101 DKR of Acyloins 101 Acid-Catalyzed Racemization 101 Racemization through Continuous Reversible Formation–Cleavage of the Substrate 102 DKR of Cyanohydrins 103 DKR of Hemithioathetals 103 Racemization Catalyzed by Aldehydes 104 Enzyme-Catalyzed Racemization 106 Racemization through SN2 Displacement 106 Other Racemization Methods 107 DKR of 5-Hydroxy-2-(5H)-Furanones 107 DKR of Hemiaminals 107 DKR of 8-Amino-5,6,7,8-tetrahydroquinoline 108 Concluding Remarks 109 References 110

4.1 4.1.1 4.1.2 4.1.3 4.2 4.2.1 4.2.2 4.2.3 4.3 4.3.1 4.3.2 4.3.3 4.3.4 4.3.5 4.4 4.5 4.5.1 4.5.2 4.6 4.7 4.8 4.9 4.9.1 4.9.2 4.9.3 4.10

5 5.1 5.2 5.2.1 5.2.2 5.2.3 5.2.4

Deracemization and Enantioconvergent Processes 115 Nicholas J.Turner Introduction 115 Deracemization Processes 116 Cyclic Oxidation–Reduction Systems 116 Microbial Deracemization of Secondary Alcohols Using a Single Microorganism 122 Deracemization of Alcohols Using Two Enzyme/Microorganism Systems 124 Epoxides 126

VII

VIII

Contents

5.2.5 5.3 5.3.1 5.3.2 5.4

Carboxylic Acids 126 Enantioconvergent Processes 127 Epoxide Hydrolysis 128 Sulfatases 129 Conclusions and Future Prospects 130 References 130

6

Transesterification and Hydrolysis of Carboxylic Acid Derivatives, Alcohols, and Epoxides 133 Robert Chênevert, Pierre Morin, and Nicholas Pelchat Introduction and General Aspects 133 Scope 133 Reaction Conditions 133 Kinetic Resolution, Dynamic Kinetic Resolution, and Desymmetrization 134 Enantioconvergent Transformation 137 Enantioselective Biotransformations of Carboxylic Acid Derivatives 137 Ester Hydrolysis 137 Ester Alcoholysis 140 Esteriﬁcation of Carboxylic Acids 140 Lactones 142 Anhydrides 143 Hydrolysis of Nitriles 144 Hydrolysis of Amides, Lactams, and Hydantoins 146 Enantioselective Enzymatic Transformations of Alcohols 150 Hydrolysis of Epoxides 157 Conclusion 162 References 163

6.1 6.1.1 6.1.2 6.1.3 6.1.4 6.2 6.2.1 6.2.2 6.2.3 6.2.4 6.2.5 6.2.6 6.2.7 6.3 6.4 6.5

7 7.1 7.2 7.3 7.4 7.4.1 7.4.2 7.4.3 7.4.4 7.5 7.6 7.7

Aminolysis and Ammonolysis of Carboxylic Acid Derivatives 171 Vicente Gotor-Fernández and Vicente Gotor Introduction 171 Mechanism of Enzymatic Ammonolysis and Aminolysis Reactions 172 Aminolysis and Ammonolysis of Carboxylic Acids 174 Aminolysis and Ammonolysis of Esters 176 Preparation of Nonchiral Amides 176 Resolution of Esters 178 Resolution of Amines, Diamines, and Aminoalcohols 180 Desymmetrization of Diesters 184 Kinetic Resolution of Secondary Amines 185 Toward the Synthesis of b-Aminoacid Derivatives 186 Summary and Outlook 188 References 189

Contents

8 8.1 8.2 8.2.1 8.2.2 8.2.3 8.2.4 8.2.5 8.2.6 8.3 8.3.1 8.3.2 8.3.2.1 8.3.2.2 8.3.2.3 8.3.2.4 8.3.3 8.3.4 8.3.4.1 8.3.4.2 8.3.4.3 8.4 8.4.1 8.4.1.1 8.4.1.2 8.4.2 8.4.3 8.4.4 8.5 8.5.1 8.5.2 8.5.3 8.5.3.1 8.5.3.2 8.6

9 9.1 9.2 9.2.1 9.2.2

Enzymatic Reduction Reaction 193 Kaoru Nakamura and Tomoko Matsuda Introduction 193 Hydrogen Source for Coenzyme Regeneration 193 Alcohol as a Hydrogen Source of Reduction 194 Sugar as a Hydrogen Source of Reduction 194 Formate as a Hydrogen Source of Reduction 194 Molecular Hydrogen as a Hydrogen Source of Reduction 195 Light Energy as a Hydrogen Source of Reduction 196 Electric Power as a Hydrogen Source of Reduction 198 Methodology for Stereochemical Control 199 Screening of Microbes 199 Modiﬁcation of Biocatalysts by Genetic Methods 201 Modiﬁed Yeast 201 Overexpression 202 Coexpression of Genes for Carbonyl Reductase and Cofactor-Regenerating Enzymes 203 Modiﬁcation of Biocatalysts: Directed Evolution 204 Modiﬁcation of Substrates 205 Modiﬁcation of Reaction Conditions 206 Acetone Treatment of the Cell 206 Selective Inhibitor 207 Reaction Temperature 208 Medium Engineering 209 Organic Solvent 209 Soluble Organic Solvent 209 Aqueous-organic Two-phase Reaction 209 Use of Hydrophobic Resin, XAD 211 Supercritical Carbon Dioxide 213 Ionic Liquid 215 Synthetic Applications 216 Reduction of Aldehyde 216 Reduction of Ketone 216 Dynamic Kinetic Resolution and Deracemization 221 Dynamic Kinetic Resolution 221 Deracemization through Oxidation and Reduction 223 Conclusions 224 References 225 Biooxidations in Chiral Synthesis 229 Marko D. Mihovilovic and Dario A. Bianchi Introduction 229 Oxidations of Alcohols and Amines 231 Regioselective Oxidation of Alcohols 231 Desymmetrizations of Diols 233

IX

X

Contents

9.2.3 9.2.4 9.2.5 9.3 9.3.1 9.3.2 9.3.3 9.4 9.5 9.5.1 9.5.2 9.5.3 9.5.4 9.6 9.6.1 9.6.2 9.6.3 9.7 9.7.1 9.7.2 9.8 9.9

10 10.1 10.2 10.3 10.3.1 10.3.2 10.4 10.4.1 10.4.2 10.4.3 10.5 10.6. 10.7 10.8

Kinetic Resolution of Primary and Secondary Alcohols Deracemization of Secondary Alcohols 235 Deracemization of Amines 237 Oxygenation of Nonactivated Carbon Centers 237 Hydroxylations by Wild-Type Whole-Cells 238 Hydroxylations by Recombinant Cytochrome P450 Monooxygenases 239 Hydroxylation via Hydroperoxide Formation 241 Enzymatic Epoxidation 241 Baeyer–Villiger Oxidations 243 Chemoselectivity 245 Desymmetrizations 246 Kinetic Resolutions 248 Regiodivergent Biooxidations 251 Heteroatom Oxidations 253 Sulﬁde Oxidation 253 Oxidation of Dithio-Compounds 256 Oxidation of other Heteroatoms 256 Aryl Dihydroxylations 257 Dioxygenation in ortho- and meta-Position 257 Dioxygenation in ipso- and ortho-Position 262 Halogenation Reactions 263 Summary and Outlook 265 References 265 Aldolases: Enzymes for Making and Breaking C–C Bonds Wolf-Dieter Fessner Introduction 275 Mechanistic Classiﬁcation 276 Pyruvate Aldolases 278 N-Acetylneuraminic Acid Aldolase 278 Related Pyruvate Aldolases 281 Dihydroxyacetone Phosphate Aldolases 284 Fructose 1,6-Bisphosphate Aldolase 285 Related DHAP Aldolases 286 Preparative Applications 288 Transketolase and Related Enzymes 302 2-Deoxy-D-Ribose 5-Phosphate Aldolase 305 Glycine Aldolases 308 Summary 311 References 311 Index

319

234

275

XI

Preface The increasing interest for obtaining chiral enantiopure organic molecules has tremendously expanded the development of new technologies and synthetic procedures pursuing this goal. Besides, a need for higher sensibility in developing environmentally friendly and sustainable processes is mandatory for the science and technology of the 21st century. Enzymatic processes combine both aspects with a high degree of success, being able to solve conceptual and practical problems. These are the main reasons for the increasing use of biotransformations both in academia and industry. The recent developments in molecular biology and biochemical tools have expanded the real applications of biotransformations, even on very large scales, and has broad applications in pharmaceutical, alimentary and cosmetic industries, just to cite a few. Enzymes work under very mild reaction conditions with a high degree of efﬁciency and selectivity. These facts constitute the foundations of an already well documented research ﬁeld with very solid bases, but continuously expanding thanks to new techniques and methodologies. For this reason, we felt necessary to collect and organize these different bio-catalytic approaches to the synthesis of enantioenriched, ideally enantiopure organic compounds. The book discusses most of the existing biocatalytic solutions to real synthetic problems, which makes it particularly helpful for synthetic, ﬁne-chemicals and pharmaceutical chemists. In all chapters our goal has been to be comprehensive and clear enough in order to also make it a friendly-to-use source of information for undergraduate and PhD students aiming to enter into this fascinating and useful research area. Thus, we believe that this book can be both a tool for studying and a bench-guide for solutions to real problems. The book has been divided in two main parts: one about methodology and a second one about synthetic applications. The ﬁrst methodological part comprises three chapters focused on the different ways of improving the stereochemical outcome of a given biotransformation. The ﬁrst chapter describes the different strategies available to improve the stereochemical outcome of a biocatalyst by modiﬁying the reaction medium. The second chapter summarises the improvement of the enzymatic selectivity by generating new catalysts through directed evolution

Asymmetric Organic Synthesis with Enzymes. Edited by Vicente Gotor, Ignacio Alfonso, and Eduardo García-Urdiales Copyright Ó 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 978-3-527-31825-4

XII

Preface

techniques. Finally, a third chapter describing different methods available to ﬁnd new enzymes with improved properties fulﬁlls the methodology part of the book. The second and larger part has been organized attending to the type of reaction catalyzed by the enzyme or the conceptually special approach. Thus, the fourth chapter deals with dynamic kinetic resolutions, an elegant and attractive way of overcoming the 50% maximum theoretical yield of classical resolutions. Two other conceptually interesting and useful approaches, such as deracemization and enantioconvergent processes, are reviewed in chapter ﬁve. The sixth chapter presents the most recent uses of enzymes in the transesteriﬁcation and hydrolysis of carboxylic acid derivatives, alcohols and epoxides, which are among the most widely used applications of enzymes in organic synthesis. In chapter seven, we have tried to compile the most recent advances in using enzymes for the resolution of amines and amide synthesis. Chapters eight and nine deal with reduction and oxidation reactions, respectively, speciﬁcally with the goal of obtaining enantioenriched interesting chiral compounds. Additionally, some intelligent solutions for the regeneration of the cofactors needed for the processes are specially highlighted. Finally, the formation of carbon-carbon bonds using enzymes is reviewed in the last chapter of the book. We would also like to thank very warmly all the chapter authors who have felt the importance of producing a book with these characteristics. They clearly understood the philosophy of the project from the beginning, and they have contributed with exceptionally well-written pieces of work in all senses. We really would like to thank them for their highly enthusiastic dedication. We must say that it has been a real pleasure to collaborate with such an excellent group of scientist from all over the world. And last but not least, we would also like to acknowledge the patience of our families and co-workers who, even without participating in this book, have suffered from the time and dedication devoted to this goal.

XIII

List of Contributors Jan-E. Bäckvall Stockholm University SE-106 Department of Organic Chemistry Arrhenius Laboratory 91 Stockholm Sweden

Wolf-Dieter Fessner Technische Universität Darmstadt Institut für Chemie und Biochemie Petersenstraße 22 64287 Darmstadt Germany

Dario A. Bianchi Vienna University of Technology Institute of Applied Synthetic Chemistry Getreidemarkt 9/163-OC 1060 Vienna Austria

Vicente Gotor Universidad de Oviedo Facultad de Química Departamento de Química Orgánica e Inorgánica Julian Clavería 6 33006 Oviedo Spain

Giacomo Carrea Istituto di Chimica del Riconoscimento Molecolare, CNR Via Mario Bianco 9 20131 Milano Italy Robert Chênevert Université Laval Faculté des Sciences et de Génie Departement de Chemie.CREFSIP. Québec, G1K 7P4 Canada

Vicente Gotor-Fernández Universidad de Oviedo Facultad de Química Departamento de Química Orgánica e Inorgánica Julian Clavería 6 33006 Oviedo Spain Romas J. Kazlauskas University of Minnesota 1479 Gortner Avenue Saint Paul, MN 55108 USA

Asymmetric Organic Synthesis with Enzymes. Edited by Vicente Gotor, Ignacio Alfonso, and Eduardo García-Urdiales Copyright Ó 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 978-3-527-31825-4

XIV

List of Contributors

Tomoko Matsuda Tokyo Institute of Technology Department of Bioengineering 4259 Nagatsuta, Midori-ku Yokohama 226-8501 Japan

Nicholas Pelchat Université Laval Faculté des Sciences et de Génie Departement de Chemie.CREFSIP. Québec, G1K 7P4 Canada

Belén Martin-Matute Departamento de Química Orgánica Universidad Autónoma de Madrid Cantoblanco 28049 Madrid Spain

Manfred T. Reetz Max-Planck-Institut für Kohlenforschung Kaiser-Wilhelm-Platz 1 45470 Mülheim an der Ruhr Germany

Stockholm University SE-106 Department of Organic Chemistry Arrhenius Laboratory 91 Stockholm Sweden

Jean-Louis Reymond University of Bern Department of Chemistry & Biochemistry Freiestraße 3 3012 Bern Switzerland

Marko D. Mihovilovic Vienna University of Technology Institute of Applied Synthetic Chemistry Getreidemarkt 9/163-OC 1060 Vienna Austria Pierre Morin Université Laval Faculté des Sciences et de Génie Departement de Chemie.CREFSIP. Québec, G1K 7P4 Canada Kaoru Nakamura Kyoto University Uji Institute for Chemical Research Kyoto 611-0011 Japan

Sergio Riva Istituto di Chimica del Riconoscimento Molecolare, CNR Via Mario Bianco 9 20131 Milano Italy Wolfgang Streit University of Bern Department of Chemistry & Biochemistry Freiestraße 3 3012 Bern Switzerland Nicholas J. Turner The University of Edinburgh Department of Chemistry Kings Buildings West Mains Road Edinburgh EH9 3JJ United Kingdom

j1

I Methodology

Asymmetric Organic Synthesis with Enzymes. Edited by Vicente Gotor, Ignacio Alfonso, and Eduardo Garcıa-Urdiales Copyright 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 978-3-527-31825-4

j3

1 Medium Engineering Giacomo Carrea and Sergio Riva

1.1 Introduction

Efﬁciency and selectivity are the two keywords that better outline the outstanding performances of enzymes. However, in some cases unsatisfactory stereoselectivity of enzymes can be found and, in these cases, the enantiomeric excesses of products are too low for synthetic purposes. In order to overcome this limitation, a number of techniques have been proposed to enhance the selectivity of a given biocatalyst. The net effect pursued by all these protocols is the increase of the difference in activation energy (DDG6¼) of the two competing diastereomeric enzyme–substrate transition state complexes (Figure 1.1). The enantioselectivity of biocatalytic reactions is normally expressed as the enantiomeric ratio or the E value [1a], a biochemical constant intrinsic to each enzyme that, contrary to enantiomeric excess, is independent of the extent of conversion. In an enzymatic resolution of a racemic substrate, the E value can be considered equal to the ratio of the rates of reaction for the two enantiomers, when the conversion is close to zero. More precisely, the E value is deﬁned as the ratio between the speciﬁcity constants (kcat/KM) for the two enantiomers and can be obtained by determination of the kcat and KM of a given enzyme for the two individual enantiomers. However, considering practical limitations, that is, the availability of optically pure enantiomers, E values are more commonly determined on racemates by evaluating the enantiomeric excess values as a function of the extent of conversion in batch reactions. For irreversible reactions, the E value can be calculated from Equation 1 (when the enantiomeric excess of the product is known) or from Equation 2 (when the enantiomeric excess of the substrate is known) [1a]. For reversible reactions, which may be the case in enzymatic resolution carried out in organic solvents (especially at extents of conversion higher than 40%), Equations 3 or 4, in which the reaction equilibrium constant has been introduced, should be used [1b].

Asymmetric Organic Synthesis with Enzymes. Edited by Vicente Gotor, Ignacio Alfonso, and Eduardo Garcıa-Urdiales Copyright 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 978-3-527-31825-4

j 1 Medium Engineering

4

[EA]

E+P

[EB]

E+Q

+A E +B

[EA]≠

ΔG

≠

ΔΔG≠

[EB]

E + A or B

E+P

E+Q

Figure 1.1 Energy diagram for an enzyme-catalyzed enantioselective reaction. E ¼ enzyme; A and B ¼ enantiomeric substrates; P and Q ¼ enantiomeric products; [EA] and [EB] ¼ enzyme–substrate complexes; DDG6¼ ¼ difference in free energy; 6¼ denotes a transition state.

For obtaining both the product and the remaining substrate in high enantiomeric excess inone reaction step, the E-value needs to be high, usually around or more than100. E¼

ln ½1cð1 þ eep Þ ln ½1cð1eep Þ

ð1Þ

E¼

ln ½ð1cÞð1ees Þ ln ½ð1cÞð1 þ ees Þ

ð2Þ

E¼

ln ½1ð1 þ KÞcð1 þ eep Þ ln ½1ð1 þ KÞcð1eep Þ

ð3Þ

E¼

ln f1ð1 þ KÞ½c þ ees ð1cÞg ln f1ð1 þ KÞ½cees ð1cÞg

ð4Þ

Where c is conversion of substrate, eep and ees are enantiomeric excesses of product (P) and remaining substrate (S), K ¼ (1 ceq)/ceq, and ceq ¼ conversion at equilibrium.

1.2 Modulation of Enzyme Enantioselectivity by Medium Engineering

Owing to the logarithmic relationship between E and DDG6¼, a small increase in DDG6¼ produces a dramatic change in E. For instance, when DDG6¼ is increased by only 1 kcal mol1 approximately, the enantiomeric excess of the product is enhanced from 80 to 95% [2]. DDG„ ¼ RTlnE

ð5Þ

The stereochemical outcome of a biocatalyzed reaction is predetermined by the enzyme, the reactant substrates, and also by the reaction conditions (temperature, pH, solvent, additives, etc.). Accordingly, each of them offers a possibility to modulate the enzymatic performances. In this chapter we are going to focus on the latter. With this regard, signiﬁcant literature examples have shown how a careful tuning of the chemical structure of the substrates as well as of the reaction conditions (temperature and pH) might help to reach the synthetic goals in terms of the enantiomeric excesses of the products (see, for instance, Ref. [3,4]). However, the parameter that has been mostly investigated is the chemical composition of the reaction medium, to the point that the so-called medium engineering has emerged in the last years as an important area of research and biotechnological development for industrial applications [ for previous reviews see 5]. In this chapter, experimental evidences of the inﬂuence of this parameter on enzymatic enantioselectivity are discussed. As will be shown in the following text, most of the literature data are related to the exploitation of hydrolases (lipases, esterases, proteases); however, some signiﬁcant examples related to other classes of enzymes have also been reported.

1.2 Modulation of Enzyme Enantioselectivity by Medium Engineering

The term medium engineering, that is the possibility to affect enzyme selectivity simply by changing the solvent in which the reaction is carried out, was coined by Klibanov, who indicated it as an alternative or an integration to protein engineering [5a]. Indeed, several authors have conﬁrmed that the enantio-, prochiral-, and even regioselectivity of enzymes can be inﬂuenced, sometimes very remarkably, by the nature of the organic solvent used. In the following text, examples of solvent effects on enzyme selectivity, referred either to systems based (i) on water-miscible organic cosolvents added to aqueous buffers or (ii) on organic media with low water activity, are discussed. 1.2.1 Selectivity Enhancement by Addition of Water-Miscible Organic Cosolvents

Initially, biocatalysis was being conducted in neat aqueous solutions because of the general notion that this environment is optimal for maintaining the enzyme conformation most suitable for binding and catalysis. However, because of the limited water-solubility of many organic substrates, it has been suggested to add varying proportions of water-miscible organic cosolvents to achieve an enhanced concentration

j5

j 1 Medium Engineering

6

R

Pig liver esterase , H2O (pH 7) MeOOC

Organic cosolvent

COOMe

HOOC

1

COOMe 2

Scheme 1.1 Table 1.1 Influence of cosolvents on the asymmetric hydrolysis of

the prochiral diester (1) catalyzed by pig liver esterase. Organic cosolvent, v/v

Temperature ( C)

Enantiomeric excesses of (2)

None Acetonitrile, 5% Acetone, 5% Dimethylsulfoxide, 5% Methanol, 5% Methanol, 20%

20 20 20 20 20 10

79 70 72 81 88 97

of the substrate in the reaction medium [6]. Quite soon, however, it became clear that these cosolvents can also inﬂuence the stereoselectivity of the biocatalysts. This area of research was pioneered by Bryan Jones and coworkers in the late 1970s and 1980s [7]. For instance, in the 1980s they showed that the pig liver esterase (PLE)catalyzed hydrolysis of the model prochiral compound 3-methyl-glutarate dimethyl ester (1) to give the (R)-monoester (2) was signiﬁcantly effected by the added cosolvent (Scheme 1.1 and Table 1.1 )[7b], the higher enantiomeric excess value of (2) being obtained in the presence of 20% v/v MeOH and at low temperature. In another signiﬁcant example published in the same period, Guanti and coworkers described the desymmetrization of the meso cyclohexene diester (3) to give the hemiester (4), whose enantiomeric excess increased from 55 to 96% when moving from plain buffer to the same buffer containing 10% v/v t-BuOH (Scheme 1.2 and Table 1.2 [8]). Finally, as an old example of kinetic resolution of racemic mixtures, mention must be made on the report of Kise and Tomiuchi on the signiﬁcant effect of acetonitrile on the enantioselectivity of different proteases toward the kinetic resolution of aromatic amino acid ethyl esters (5–8). For instance, (L)-DOPA (8) was obtained with 99% ee in the presence of 90% v/v acetonitrile [9]. COOR¢¢

R

NHR¢

HO

OAc OAc 3 Scheme 1.2 .

5 6 7 8

R = R¢ = H ; R¢¢ = Et R = H ; R¢ = Ac ; R¢¢ = Et R = OH ; R¢ = H ; R¢¢ = Et R = OH ; R¢ = H ; R¢¢ = H

Pig liver esterase , H2O (pH 7)

OH OAc

Organic cosolvent 4

1.2 Modulation of Enzyme Enantioselectivity by Medium Engineering Table 1.2 Influence of cosolvents on the asymmetric hydrolysis of the meso-diester (3) catalyzed by pig liver esterase.

Organic cosolvent, v/v

Relative rate of hydrolysis

Enantiomeric excesses of (4)

None Dimethylsulfoxide, 20% Dimethylsulfoxide, 40% Dimethylformamide, 20% tert-Butanol, 5% tert-Butanol, 10%

100 70 28 35 70 44

55 59 72 84 94 96

In a very recent report related to a different group of enzymes, it has been shown that cosolvents, especially methanol, can improve to a great extent the enantioselectivity of three Baeyer–Villiger monooxygenases (phenylacetone monooxygenase, 4-hydroxyacetophenone monooxygenase, and ethionamide monooxygenase) when using organic sulﬁdes as the substrates; depending on the enzyme and on the nature of the solvent and the substrate used, reversal of enantiopreference was also observed [10]. Several other examples have been reported in the literature and most of them have been already reviewed (see, for instance, [11]). However, it must be mentioned that, of course, the selectivity enhancement via addition of water-miscible organic cosolvents may not be taken for granted, as sometimes this approach may be unsuccessful [11a]. The discussion on possible rationales of this phenomenon has been reported at the end of this chapter. 1.2.2 Selectivity Enhancement in Organic Media with Low Water Activity 1.2.2.1 Organic Solvent Systems In the third example of the previous paragraph [9], the reaction conditions described are similar to those used for the enzymatic transformations in bulk organic media, an area that was pioneered by Klibanov and coworkers in the 1980s [12] and later investigated and synthetically exploited worldwide [13]. In these systems, solid enzyme preparations (e.g. lyophilized or immobilized on a support) are suspended in an organic solvent in the presence of enough aqueous buffers to ensure catalytic activity. Although the amount of water added to the solvent (as a rule of thumb <5% v/v) may exceed its solubility in that solvent, a visible discrete aqueous phase is not apparent because part of it is adsorbed by the enzyme. Therefore, the two phases involved in an organic solvent system are a liquid (bulk organic solvent and reagents dissolved in it) and a solid (hydrated enzyme particles). The interest and success of the enzyme-catalyzed reactions in this kind of media is due to several advantages such as (i) solubilization of hydrophobic substrates; (ii) ease of recovery of some products; (iii) catalysis of reactions that are unfavorable in water (e.g. reversal of hydrolysis reactions in favor of synthesis); (iv) ease of recovery of insoluble biocatalysts; (v) increased biocatalyst thermostability; (vi) suppression of water-induced side reactions. Furthermore, as already said, enzyme selectivity can be markedly inﬂuenced, and even reversed, by the solvent.

j7

j 1 Medium Engineering

8

However, in most cases enzymes show lower activity in organic media than in water. This behavior has been ascribed to different causes such as diffusional limitations, high saturating substrate concentrations, restricted protein ﬂexibility, low stabilization of the enzyme–substrate intermediate, partial enzyme denaturation by lyophilization that becomes irreversible in anhydrous organic media, and, last but not least, nonoptimal hydration of the biocatalyst [12d]. Numerous methods have been developed to activate enzymes for optimal use in organic media [13]. Before discussing the medium engineering phenomenon and its synthetic relevance in details, it is useful to offer a brief overview of the fundamentals of biocatalysis in organic media. 1.2.2.2 Enzyme Properties in Organic Solvents Enzyme activity in organic solvents depends on parameters such as water activity, pH control, substrate–product solvation, enzyme form, and nature of the solvent. Water Activity It became apparent quite soon that enzyme activity was higher in hydrophobic solvents than in hydrophilic ones and, indeed, Laane et al. [14] found that there was a direct correlation between activity and solvent hydrophobicity, expressed as log P (where P isthepartitioncoefﬁcient of a givensolventbetweenn-octanol and water). Soon after, Zaks and Klibanov [12c] demonstrated, with three unrelated oxidoreductases, that enzyme activity was more or less coincident in the various solvents, provided that the amount of water bound to the protein was the same. The maximal enzymatic activity for the three examined enzymes was attained at about 1000 molecules of water/ enzyme molecule, that is, roughly a monolayer of water on the surface. A more rigorous way to correlate enzyme activity with the water present in the reaction medium was proposed by Halling [15] and Goderis [16], who expressed the water in the medium no longer in terms of content but in terms of thermodynamic water activity (aw). Water activity is correlated to mole fraction of water (ww) and water activity coefﬁcient (gw) by the equation aw ¼ wwgw . Since, in ﬁrst approximation, the water activity coefﬁcient (gw) increases as a function of solvent hydrophobicity, it follows that a given value of aw will be obtained at a lower water concentration in a hydrophobic than in a hydrophilic medium. Therefore, to attain the same level of enzyme hydration, or of aw, in the enzyme, less water is necessary in hydrophobic than in hydrophilic solvents. Because of the fundamental role played by water, it is clear that the control of aw in the enzymatic reactions carried out in organic solvents is of paramount importance and, as a consequence, different methods have been developed to this end, all of them based on the knowledge that some salts have a ﬁxed hydration state, which gives a well-deﬁned aw [17]. pH and pH Control It is well known that the protonation state of the various groups of an enzyme is important for enzyme activity. However, while in water protonation is simply controlled by pH adjustments, this is not the case in organic solvents, where even the pH concept itself is challenged. A simple way to control the initial state of enzyme protonation was developed by Zaks and Klibanov [12a] who showed that enzyme activity in organic solvents was markedly dependent on the pH of the solution from which the enzyme was recovered (by lyophilization or precipitation).

1.2 Modulation of Enzyme Enantioselectivity by Medium Engineering

This methodology, the so-called pH adjustment, is now widely applied and is perfectly suitable when there is no alteration of acid–base concentration in the course of the reaction. If this is not the case, as, for instance, in the synthesis, hydrolysis, or aminolysis of esters, enzyme protonation and activity might be deeply inﬂuenced. To overcome this drawback, the use of buffering systems involving highly hydrophobic acids and their sodium salts, and highly hydrophobic bases and their hydrochlorides, has been proposed [18]. The main disadvantage of these buffer systems is that they are appreciably soluble only in solvents with a relatively high polarity. Enzyme Form Proteins are practically insoluble in most organic solvents; therefore, in the absence of any special treatment, they are usually present as a solid suspension. This simpliﬁes catalyst–product separation and enzyme reutilization. The simplest way to prepare a biocatalyst for use in organic solvents and, at the same time, to adjust key parameters, such as pH, is its lyophilization or precipitation from aqueous solutions. These preparations, however, can undergo substrate diffusion limitations or prevent enzyme–substrate interaction because of protein–protein stacking. Enzyme lyophilization in the presence of lyoprotectants (polyethylene glycol, various sugars), ligands, and salts have often yielded preparations that are markedly more active than those obtained in the absence of additives [19]. Besides that, the addition of these ligands can also affect enzyme selectivity as follows. Adsorption on solid matrices, which improves (at optimal protein/support ratios) enzyme dispersion, reduces diffusion limitations and favors substrate access to individual enzyme molecules. Immobilized lipases with excellent activity and stability were obtained by entrapping the enzymes in hydrophobic sol-gel materials [20]. Finally, in order to minimize substrate diffusion limitations and maximize enzyme dispersion, various approaches have been attempted to solubilize the biocatalysts in organic solvents. The most widespread method is the one based on the covalent linking of the amphiphilic polymer polyethylene glycol (PEG) to enzyme molecules [21]. Enzyme Kinetics and Stability Kinetic studies, carried out mostly with hydrolases, have shown that enzymes in organic solvents follow conventional models [12a, 22]. Several reports have indicated that enzymes are more thermostable in organic solvents than in water. The high thermal stability of enzymes in organic solvents, especially in hydrophobic ones and at low water content, was attributed to increased conformational rigidity and to the absence of nearly all the covalent reactions causing irreversible thermoinactivation in water [23]. 1.2.2.3 Medium Engineering In the following text, a few examples of solvent effects on enzyme selectivity are discussed, starting from the classical works published by Klibanovs group. In a ﬁrst report [24], the enantioselectivities of various proteases were evaluated by comparing the biocatalyzedhydrolysisof 2-chloroethyl esters of N-acetyl-L- and D-amino acids in water and their transesteriﬁcation with n-propanol in butyl ether. By comparing the ratio of the kcat/KM values for the L- and D-enantiomers in the two reactions, a remarkable relation of the proteases enantioselectivity was observed: apparently, in this case, the organic solvents destroyed the selectivity of the tested enzymes. This ﬁnding

j9

j 1 Medium Engineering

10

Table 1.3 Influence of the organic solvent on the enantioselectivity of the protease subtilisin in the kinetic resolution of the racemic amine (9) (expressed as the ratio of the initial rate of acylation of the pure enatiomers, vS/vR).

Organic solvent

Selectivity (vS/vR)

Toluene Octane Acetonitrile Carbon tetrachloride Ethyl acetate Butyl ether Pyridine Dimethylformamide Tetrahydrofuran tert-Amyl alcohol 3-Methyl-3-pentanol

0.95 1.3 1.4 1.5 1.6 1.8 2.5 2.9 3.5 4.1 7.7

was synthetically exploited for the preparation of peptides containing D-amino acids, compounds that, in water, could not be substrates for these enzymes [25]. Later on the crucial role played by the solvent was enlightened in the proteasecatalyzed resolution of racemic amines [26]. As shown in Table 1.3, the ratio of the initial rates of acylation of the (S)- and the (R)-enantiomers or racemic a-methylbenzylamine (9) varied from nearly 1 in toluene to 7.7 in 3-methyl-3-pentanol. Similarly, the same authors found a signiﬁcant solvent effect for the subtilisincatalyzed transesteriﬁcation of racemic 1-phenylethanol (10) using vinyl butyrate as acyl donor (Table 1.4 [27]). NH2

9

OH

10

Table 1.4 Influence of the organic solvent on the enantioselectivity of the protease subtilisin in the kinetic resolution of the racemic alcohol (10) (expressed as the enatiomeric ratio E, that is the ratio of the specificity constants of the two enatiomers, (Kcat/KM)S/ (Kcat/KM)R).

Organic solvent

(Kcat/KM)S

(Kcat/KM)R

E

Dioxane Benzene Triethylamine Tetrahydrofuran Pyridine Dimethylformamide Nitromethane Acetonitrile

170 13 330 230 43 1.4 16 48

2.8 0.24 6.9 5.8 1.4 0.16 3.3 16

61 54 48 40 31 9 5 3

1.2 Modulation of Enzyme Enantioselectivity by Medium Engineering

Of course, the inﬂuence of organic solvents on enzyme enantioselectivity is not limited to proteases but it is a general phenomenon. Quite soon, different research groups described the results obtained with lipases [28]. For instance, the resolution of the mucolytic drug ()–trans-sobrerol (11) was achieved by transesteriﬁcation with vinyl acetate catalyzed by the lipase from Pseudomonas cepacia adsorbed on celite in various solvents. As depicted in Scheme 1.3 and Table 1.5, it was found that t-amyl alcohol was the solvent of choice: in this medium, the selectivity was so high (E > 500) that the reaction stopped spontaneously at 50% conversion giving both (þ)-transsobrerol and ()-trans-sobrerol monoacetate in 100% optical purity [29]. O

HO Lipase PS ,

AcO

HO

O

+

Organic solvent OH

OH

OH

(±)-trans -11

(1S, 5R)-12

(1R, 5S)-11

Scheme 1.3 . Table 1.5 Influence of the organic solvent on the enantioselectivity of the lipase PS (from Pseudomonas species) in the kinetic resolution of racemic trans-sobrerol (10).

Organic solvent

E

Tetrahydrofuran Acetone Dioxane 3-Pentanone tert-Amyl alcohol

69 142 178 212 518

In all the reported examples, the enzyme selectivity was affected by the solvent used, but the stereochemical preference remained the same. However, in some speciﬁc cases it was found that it was also possible to invert the hydrolases enantioselectivity. The ﬁrst report was again from Klibanovs group, which described the transesteriﬁcation of the model compound (13) with n-propanol. As shown in Table 1.6, the enantiopreference of an Aspergillus oryzae protease shifted from the (L)- to the (D)-enantiomer by moving from acetonitrile to CCl4 [30]. Similar observations on the inversion of enantioselectivity by switching from one solvent to another were later reported by other authors [31]. O Cl

O NHAc

13

When discussing the role of reaction medium on enzyme enantioselectivity, the potential effects of (i) water activity [5b,13f,32], (ii) enzyme form, and (iii) pH, should

j11

j 1 Medium Engineering

12

Table 1.6 Influence of the organic solvent on the enantioselectivity of the protease from A. oryzae subtilisin in the kinetic resolution of the racemic amino acid (12) (expressed as the ratio of the initial rate of acylation of the pure enatiomers, vS/vR).

Organic solvent

Selectivity (vS/vR)

Acetonitrile Dimethylformamide Pyridine Tert-Butanol Acetone Tetrahydrofuran Cycloexanone Methylene chloride 3-Octanone Nitrobenzene tert-Butyl acetate Methyl tert-butyl ether Cyclohexane Toluene Octane Carbon tetrachloride

7.1 5.7 4.3 1.7 1.3 1.3 1.1 0.88 0.73 0.60 0.44 0.34 0.27 0.26 0.24 0.19

not be neglected. As for the ﬁrst point, the best way to avoid water effects is to carry out experiments at controlled water activity, which indeed has been done in numerous cases [ for a review see [5b,13a]]. When the effects of water activity (or water content) have been speciﬁcally investigated, the results were rather contradictory since increases, decreases, or, most often, no variation in the enantioselectivity as a function of the water present in the reaction medium were observed [5b,13e,33,34]. Contradictory results were also obtained for enzyme form, since no inﬂuence [35] or inﬂuence [36] on enzyme enantioselectivity was reported. In the latter case, the large differences in stereoselectivity as a function of Candida antarctica lipase B form were ascribed to mass transport limitations, with lyophilized preparations more prone to this problem than immobilized preparations [36]. Considering pH, it has been demonstrated that alcohol dehydrogenase enantioselectivity is affected by pH variations in aqueous buffers [37]. However, a study carried out in organic solvents showed that the hydrolytic enzyme cutinase was not inﬂuenced by pH changes [34]. 1.2.3 Rationales

The experimental evidences that medium engineering might represent an efﬁcient method to modify or improve enzyme selectivity (alternative to protein engineering and to the time-consuming search for new catalysts) were immediately matched by the search for a sound rationale of this phenomenon. The different hypotheses formulated to try to rationalize the effects of the solvent on enzymatic enantioselectivity can be grouped into three different classes. The ﬁrst hypothesis suggests that

1.2 Modulation of Enzyme Enantioselectivity by Medium Engineering

the solvent, depending on its polarity, could modify the biocatalyst conformation and, thus, inﬂuence the selectivity by altering the molecular recognition process between substrate and enzyme [27]. The second group of studies tries to explain the solvent effects on enantioselectivity by means of the contribution of substrate solvation to the energetics of the reaction [38]. For instance, a theoretical model based on the thermodynamics of substrate solvation was developed [39]. However, this model, based on the determination of the desolvated portion of the substrate transition state by molecular modeling and on the calculation of the activity coefﬁcient by UNIFAC, gave contradictory results. In fact, it was successful in predicting solvent effects on the enantio- and prochiral selectivity of g-chymotrypsin with racemic 3-hydroxy-2-phenylpropionate and 2-substituted 1,3propanediols [39], whereas it failed in the case of subtilisin and racemic sec-phenetyl alcohol and trans-sobrerol [40]. That substrate solvation by the solvent can contribute to enzyme enantioselectivity was also claimed in the case of subtilisin-catalyzed resolution of secondary alcohols [41]. In a third model it has been proposed that solvent molecules could bind within the active site and, depending on their structure, interfere with the association or transformation of one enantiomer more than the other one [28a,42,43]. For instance, a correlation between the enantiomeric ratio E and the van der Waals volume of the solvent molecules was observed in the resolution of 3-methyl-2-butanol catalyzed by C. antarctica lipase B; the van der Waals volume was suggested as one of the parameters that govern solvents effects on enzyme selectivity [44]. Additionally, signiﬁcant evidence at a molecular level came from a series of crystallographic data reported by Klibanov [45]. In these works, whose main goal was to demonstrate that the tertiary structure of enzymes was the same once the proteins had been dissolved in water or suspended in organic solvents, it was shown that molecules of the organic solvents were able to penetrate into the active site of the protease subtilisin, displacing some of the water molecules present. Quite signiﬁcantly, molecules of different solvents were found to be located in different regions of the active site and, additionally, this happened not only in pure organic solvents but also in mixtures of organic solvents and water (i.e. dioxane–water 40 : 60 v/v) [45b]. On the basis of these results, it is not surprising that the search for an exhaustive rationale for the medium engineering effect is still the object of scientiﬁc debate. Furthermore, it should be pointed out that despite the aforementioned experimental results supporting the second and third type of hypotheses, both of them lack fully reliable predictive value and are not sufﬁcient by themselves to explain every case. In fact, the second hypothesis appears to be valid only for the speciﬁc enzyme and substrate investigated each time, or when the formation of solvent–enzyme complexes is proposed (third hypothesis), no generalization is possible because of the large number of possible solvent–enzyme complexes and because each complex might behave differently depending on the nature of the substrate. However, whatever the mechanism of action is, the effect of solvents on enzyme selectivity is sometimes really dramatic. For example, Hirose et al. [42] reported that in the Pseudomonas species lipase-catalyzed desymmetrization of prochiral

j13

j 1 Medium Engineering

14

NO 2 Cyclohexane HOOC

COOR

NO 2 (R)-15

Pseudomonas sp. lipase ROOC

N H NO 2

COOR N H

14 , R = t BuC(O)OCH 2 or R = EtC(O)OCH 2

ROOC

COOH

i-Propyl ether

(S)-15

N H

Scheme 1.4 Asymmetric hydrolysis of dihydropyridine diesters: influence of solvent on lipase stereochemical preference.

dihydropyridine dicarboxylates (i.e. 14), the (S)-monoesters as high as 99% ee were obtained in water-saturated di-isopropyl ether, whereas the (R)-isomers were formed preferentially (88–91% ee) in water-saturated cyclohexane (Scheme 1.4). 1.2.4 Modulation of Enzyme Selectivity: New Trends of Research

The aspects of medium engineering summarized so far were a hot topic in biocatalysis research during the 1980s and 1990s [5]. Nowadays, all of them constitute a well-established methodology that is successfully employed by chemists in synthetic applications, both in academia and industry. In turn, the main research interests of medium engineering have moved toward the use of ionic liquids as reaction media and the employment of additives. 1.2.4.1 Ionic Liquids Ionic liquids, which can be deﬁned as salts that do not crystallize at room temperature [46], have been intensively investigated as environmentally friendly solvents because they have no vapor pressure and, in principle, can be reused more efﬁciently than conventional solvents. Ionic liquids have found wide application in organometallic catalysis as they facilitate the separation between the charged catalysts and the products. In recent years ionic liquids have also been employed as media for reactions catalyzed both by isolated enzymes and by whole cells, and excellent reviews on this topic are already available [47]. Biocatalysis has been mainly conducted in those roomtemperature ionic liquids that are composed of a 1,3-dialkylimidazolium or Nalkylpyridinium cation and a noncoordinating anion [47a]. The polarity of common ionic liquids is in the range of the lower alcohols or formamide, and their miscibility with water varies widely and unpredictably and is

1.2 Modulation of Enzyme Enantioselectivity by Medium Engineering

not strictly correlated with their polarity [47a]. Even though the miscibility of ionic liquids and organic solvents is not yet well documented, generally they mix with lower alcohols and ketones, dichloromethane, and THF, whereas they do not mix with alkanes and ethers [48]. The possibility of using solvents with high polarity, such as that of many ionic liquids, increases the solubility of polar substrates and extends the range of applications of biocatalysis. Hydrolases such as lipases (from C. antarctica and from P. cepacia), proteases (thermolysin, a-chymotrypsin), esterases, and glycosidases have been used in plain ionic liquids, whereas oxidoreductases such as formate dehydrogenase, peroxidases, laccases, and Baker yeast have been employed in waterionic liquid mixtures or in water-ionic liquid biphasic systems [47]. Both increased enzyme stability and activity have been reported as compared to conventional organic solvents media. For example, a-chymotrypsin, lipase from C. antarctica and esterase from Bacillus stearothermophilus were found to be 17, 3, and 30 times more stable in ionic liquids than in organic solvents [47b]. Several reports deal also with the effects of ionic liquids on enzyme enantioselectivity, which is the subject of this chapter. Although in several cases there was no change or even a decrease in enantioselectivity compared to organic solvents [47], in other cases improved enantioselectivity was observed [47,49–56]. In the following text, the latter cases will be examined in some detail. Lipases from C. antarctica and P. cepacia showed higher enantioselectivity in the two ionic liquids 1-ethyl-3-methylimidazolium tetraﬂuoroborate and 1-butyl-3methylimidazolium hexaﬂuoroborate than in THF and toluene, in the kinetic resolution of several secondary alcohols [49]. Similarly, with lipases from Pseudomonas species and Alcaligenes species, increased enantioselectivity was observed in the resolution of 1-phenylethanol in several ionic liquids as compared to methyl tert-butyl ether [50]. Another study has demonstrated that lipase from Candida rugosa is at least 100% more selective in 1-butyl-3-methylimidazolium hexaﬂuoroborate and 1-octyl-3nonylimidazolium hexaﬂuorophosphate than in n-hexane, in the resolution of racemic 2-chloro-propanoic acid [51]. The examples described so far refer to experiments carried out in plain ionic liquids. However, it has been proved that ionic liquids are also capable of positively inﬂuencing enzyme enantioselectivity when employed in mixtures with aqueous buffers or conventional organic solvents. For example, N-ethyl-pyridinium triﬂuoroacetate improved the selectivity of subtilisin Carlsberg in the resolution of several amino acid esters in comparison with acetonitrile; both the ionic liquid and acetonitrile were present in 15% concentration in an aqueous buffer [52]. A subsequent study demonstrated that subtilisin enantioselectivity was related to the kosmotropicity of individual cations and anions of ionic liquids and that it was higher at higher values of the overall ionic liquid kosmotropicity [53]. Ionic liquids–tertbutanol cosolvent systems markedly boosted the activity, stability, and enantioselectivity of C. antarctica lipase B in the resolution of racemic p-hydroxy-phenylglycine methyl ester [54] and, similarly, cosolvent systems consisting of ionic liquids and chloroform or tert-butanol increased the selectivity of the same enzyme in the resolution of racemic methyl mandelate [55]. Finally, the three-component system

j15

j 1 Medium Engineering

16

made of ionic liquid–isopropanol–water gave the best results in the enantioselective hydrolysis of prochiral diester malonates [56]. 1.2.4.2 Additives Basically, there are three ways to tune enzyme enantioselectivity by means of additives: (i) the additives are placed in the reaction medium together with the organic solvent, the enzyme, and the reagents; (ii) the additives are co-lyophilized with the biocatalyst before use in the organic solvent; (iii) the additives are complexed with the substrates before their transformation in the organic medium. One example that refers to the ﬁrst case is the resolution of racemic 2-phenyl-4benzyloxazol-5(4H)-one (12) by alcoholysis with n-butanol catalyzed by C. antarctica lipase B or by Mucor miehei lipase [57]. The enantioselectivity of both enzymes was improved by addition to the organic medium of an organic base such as triethylamine or of a solid-state buffer such as 3-(cyclohexylamino)-2-hydroxy-1-propanesulfonic acid and its sodium salt. The positive effect of the additives on the enantioselectivity (and activity) of the two enzymes was attributed to the control of the protonation state of the biocatalysts [54]. Similarly, the addition of a small amount of an aqueous solution containing metal ions (LiCl or MgCl2) to an organic solvent medium, remarkably enhanced (up to 200 times) the enantioselectivity of C. rugosa lipase in the resolution of racemic 2-(4-substituted phenoxy) propionic acids [58]. EPR experiments suggested that the increase in enzyme selectivity brought about by the metal ions was due to an enhancement of the reaction rate for the R-enantiomer combined with a decrease of the rate for the S-enantiomer [58]. In the second way by which additives can inﬂuence enzyme enantioselectivity, it has been shown that including excess salt (e.g. KCl) during lyophilization can enhance not only the activity of subtilisin Carlsberg by more than 20 000 fold, but also its enantioselectivity toward N-acetyl-phenylalanine methyl ester [59]. The changes in selectivity reﬂected the activity for the (L)-enantiomer, whereas the activity for the (D)-enantiomer was mostly unaffected, which suggests that the favored reaction is more sensitive to the structural integrity of the salt-treated enzyme [59]. Enhanced enantioselectivity (and activity) in organic solvents was also observed for lipase from P. cepacia co-lyophilized with crown ethers and cyclodextrins [60]. Colyophilization with cyclodextrins improved the enantioselectivity of subtilisin Carlsberg and C. rugosa lipase as well [61]. The cyclodextrin effect was ascribed to the preservation of the enzyme active site and not to the formation of additive–substrate or –product complexes [61]. Finally, co-lyophilizing horseradish peroxidase with numerous amino acids inﬂuenced enzymes subsequent stereoselectivity in the sulfoxidation of methyl phenyl sulﬁde in 2-propanol. The greatest effect was observed with D-proline, which increased enzyme selectivity from a level that was synthetically meaningless to a useful one [62]. An example that refers to the third method additives can be employed is described below. Markedly enhanced enantioselectivity was reported for P. cepacia lipase and subtilisin Carlsberg with chiral substrates converted to salts by treatment with numerous Bronsted–Lowry acids or bases [63]. This effect was observed in various organic solvents but not in water, where the salts apparently dissociate to regenerate

References

the free substrates. It should be emphasized that in some instances the enzymes exhibited virtually no enantiopreference toward a free substrate but a profound one toward its salt [63].

1.3 Conclusions and Outlooks

Nowadays biocatalysis is a well-assessed methodology that has moved from the original status of academic curiosity to become a widely exploited technique for preparative-scale reactions, up to the point that the so-called industrial biotechnology (to which biocatalysis contributes to the most extent) is one of the three pillars of the modern sustainable chemistry. As shown in this chapter, by focusing on the modulation of enzyme selectivity by medium engineering, quite simple modiﬁcations of the solvent composition can really have signiﬁcant effects on the performances of the biocatalysts. The main drawback remains the lack of reliable predictive models. Despite the signiﬁcant research efforts (particularly in the last decade), it is likely that a reasonable foresight of the enantioselective outcome of an enzymatic transformation will continue to be based solely on a careful analysis of the increasingly numerous literature reports.

References 1 (a) Chen, C.S., Fujimoto, Y., Girdaukas, G. and Sih, C.J. (1982) Journal of the American Chemical Society, 104, 7294–7298. (b) Chen, C.-S. and Sih, C.J. (1989) Angewandte Chemie International Edition in English, 28, 695–707. 2 Jones, J.B. (1990) Pure and Applied Chemistry, 62, 1445–1448. 3 Phillips, R.S. (1992) Enzyme and Microbial Technology, 14, 417–419. 4 Barton, M.J., Hamman, J.P., Fitcher, K.C. and Calton, G.J. (1990) Enzyme and Microbial Technology, 12, 577–583. 5 (a) Wescott, C.R. and Klibanov, A.M. (1994) Biochimica et Biophysica Acta, 1206, 1–9. (b) Carrea, G., Ottolina, G. and Riva, S. (1995) Trends in Biotechnology, 13, 63–70. 6 Butler, L.G. (1979) Enzyme and Microbial Technology, 1, 253–259. 7 (a) Jones, J.B. and Mehes, M.M. (1979) Canadian Journal of Chemistry, 57, 2245–2248. (b) Lam, L.K.P., Hui, R.A. and

8

9 10

11

12

Jones, J.B. (1986) The Journal of Organic Chemistry, 51, 2047–2050. Guanti, G., Banﬁ, L., Narisano, E., Riva, R. and Thea, S. (1986) Tetrahedron Letters, 27, 4639–4642. Kise, H., and Tomiuchi, Y. (1991) Biotechnology Letters, 13, 317–322. de Gonzalo, G., Ottolina, G., Zambianchi, F., Fraaije, M.W. and Carrea, G. (2006) Journal of Molecular Catalysis B: Enzymatic, 39, 91–97. (a) Faber, K., Ottolina, G. and Riva, S. (1993) Biocatalysis, 8, 91–132. (b) Santaniello, E., Ferraboschi, P., Grisenti, P. and Manzocchi, A (1992) Chemical Reviews, 92, 1071–1140. (a) Zaks, A. and Klibanov, A.M. (1985) Proceedings of the National Academy of Sciences of the United States of America, 82, 3192–3196. (b) Klibanov, A.M. (1986) Chemtech, 16, 354–359. (c) Zaks, A. and Klibanov, A.M. (1988) The Journal of

j17

j 1 Medium Engineering

18

13

14

15

16

17

18

Biological Chemistry, 263, 8017–8021. (d) For a recent review see Klibanov, A.M. (2001) Nature, 409, 241–246. (a) Carrea, G. and Riva, S. (2000) Angewandte Chemie International Edition in English, 39, 2226–2254. (b) Koskinen, A.M.P. and Klibanov, A.M. (eds) (1996) Enzymatic Reactions in Organic Media, Blackie Academic & Professional, London. (c) Hudson, E.P., Eppler, R.K. and Clark, D.S. (2005) Current Opinion in Biotechnology, 16, 637–643. (d) Gupta, M.N. and Roy, I. (2004) European Journal of Biochemistry/FEBS, 271, 2575–2583. (e) Klibanov, A.M. (2003) Current Opinion in Biotechnology, 14, 427–431. (f) Halling, P.J. (2000) Current Opinion in Chemical Biology, 4, 74–80. (g) Wolff, A., Straathof, A.J.J., Jongejan, J.A. and Heijnen, J.J. (1997) Biocatalysis and Biotransformation, 15, 175–184. Laane, C., Boeren, S., Vos, K. and Veeger, C. (1987) Biotechnology and Bioengineering, 30, 81–87. (a) Halling, P.J. (1994) Enzyme and Microbial Technology, 16, 178–206. (b) Halling, P.J. (1989) Trends in Biotechnology, 7, 50–52. (c) Halling, P.J. (1984) Enzyme and Microbial Technology, 6, 513–514. Goderis, H.L., Ampe, G., Feyten, M.P., Fouwe, B.L., Guffens, W.M., Van Cauwenbergh, S.M. and Tobback, P.P. (1987) Biotechnology and Bioengineering, 30, 258–266. (a) Wehtje, E., Jasmedh, K., Adlercreutz, P., Chand, S. and Mattiasson, B. (1997) Enzyme and Microbial Technology, 21, 502–510. (b) Dudal, Y. and Lortie, R. (1995) Biotechnology and Bioengineering, 45, 129–134. (c) Ljunger, G., Adlercreutz, P. and Mattiasson, B. (1994) Enzyme and Microbial Technology, 16, 751–755. (d) Yang, Z. and Robb, D. (1993) Biotechnology Techniques, 7, 37–42. (e) Halling, P.J. (1992) Biotechnology Techniques, 6, 271–276. (a) Xu, K. and Klibanov, A.M. (1996) Journal of the American Chemical Society, 118, 9815–9819. (b) Blackwood, A.D., Curran, L.J., Moore, B.D. and Halling, P.J.

19

20

21

22

23

(1994) Biochimica et Biophysica Acta, 1206, 161–165. (a) Mingarro, I., Gonzales-Navarro, H. and Braco, L. (1996) Biochemistry, 35, 9935–9944. (b) Triantafyllou, A.O., Wehtje, E., Adlercreutz, P. and Mattiasson, B. (1995) Biotechnology and Bioengineering, 45, 406–414. (c) Khmelnitsky, Y.L., Welch, S.H., Clark, D.S. and Dordick, J.S. (1994) Journal of the American Chemical Society, 116, 2647–2648. (d) Dabulis, K. and Klibanov, A.M. (1993) Biotechnology and Bioengineering, 41, 566–571. (e) Yamane, T., Ichiryu, T., Nagata, M., Ueno, A. and Shimizu, S. (1990) Biotechnology and Bioengineering, 36, 1063–1069. (f) Russell, A.J. and Klibanov, A.M. (1988) The Journal of Biological Chemistry, 263, 11624–11626. Reetz, M.T. and Zonta, A. (1995) Angewandte Chemie International Edition in English, 34, 301–303. (a) Bovara, R., Carrea, G., Gioacchini, A.M., Riva, S. and Secundo, F. (1997) Biotechnology and Bioengineering, 54, 50–57. (b) Matsushima, A., Kodera, Y., Hiroto, M., Nishimura, H. and Inada, Y. (1996) Journal of Molecular Catalysis B: Enzymatic, 2, 1–17 and references therein. (a) Martinelle, M. and Hult, K. (1995) Biochimica et Biophysica Acta, 1251, 191–197. (b) Chatterjee, S. and Russell, A.J. (1993) Enzyme and Microbial Technology, 15, 1022–1029. (c) Rizzi, M., Stylos, P., Riek, A. and Reuss, M. (1992) Enzyme and Microbial Technology, 14, 709–714. (a) Volkin, D.B., Staubli, A., Langer, R. and Klibanov, A.M. (1991) Biotechnology and Bioengineering, 37, 843–853. (b) GarzaRamos, G., Darszon, A., Tuena de GomezPuyou, M. and Gomez-Puyou, A. (1989) Biochemistry, 28, 3177–3182. (c) Ayala, G., Tuenade Gomez-Puyou, M., GomezPuyou, A. and Darszon, A. (1986) FEBS Letters, 203, 41–43. (d) Wheeler, C.J. and Croteau, R. (1986) Archives of Biochemistry and Biophysics, 248, 429–434. (e) Reslow, M., Adlercreutz, P. and Mattiasson, B.

References

24

25

26

27

28

29

30

31

32

33

34

(1987) Applied Microbiology and Biotechnology, 26, 1–8. (f) Zaks, A. and Klibanov, A.M. (1984) Science, 224, 1249–1251. Sakurai, T., Margolin, A.L., Russell, A.J. and Klibanov, A.M. (1988) Journal of the American Chemical Society, 110, 7236–7237. Margolin, A.L., Tai, D.F. and Klibanov, A.M. (1987) Journal of the American Chemical Society, 109, 7885–7887. Kitaguci, H., Fitzpatrick, P.A., Huber, J.E. and Klibanov, A.M. (1989) Journal of the American Chemical Society, 111, 3094–3095. Fitzpatrick, P.A. and Klibanov, A.M. (1991) Journal of the American Chemical Society, 113, 3166–3171. See, for instance: (a) Secundo, F., Riva, S. and Carrea, G. (1992) Tetrahedron: Asymmetry, 3, 267–280. (b) Gutman, A.L. and Shapira, M. (1991) Journal of the Chemical Society, Chemical Communications, 1467–1468. (c) Parida, S. and Dordick, J.S. (1991) Journal of the American Chemical Society, 113, 2253–2259. (d) Garcia, M.J., Brieva, R., Rebolledo, F. and Gotor, V. (1991) Biotechnology Letters, 13, 867–870. Bovara, R., Carrea, G., Ferrara, L. and Riva, S. (1991) Tetrahedron: Asymmetry, 2, 931–938. Tawaki, S. and Klibanov, A.M. (1992) Journal of the American Chemical Society, 1143, 1882–1884. (a) Baldoli, C., Maiorana, S., Carrea, G. and Riva, S. (1993) Tetrahedron: Asymmetry, 4, 767–772. (b) Ueji, S., Fujino, R., Okubo, N., Miyazawa, T., Kurita, S., Kitadani, M. and Muromatsu, A. (1992) Biotechnology Letters, 14, 163–168. Ducret, A., Trani, M. and Lortie, R. (1998) Enzyme and Microbial Technology, 22, 212–216. Jacobsen, E.E. and Anthonsen, T. (2002) Canadian Journal of Chemistry, 80, 577–581. Vidinha, P., Harper, N., Micaelo, N.M., Lourengo, N.M.T., da Silva, M.D.R.G.,

35

36 37

38

39

40

41

42

43

44

45

46

Cabral, J.M.S., Afonso, C.A.M., Soares, C.M. and Barreiros, S. (2004) Biotechnology and Bioengineering, 85, 442–449. Ottolina, G., Carrea, G., Riva, S., Sartore, L. and Veronese, F.M. (1992) Biotechnology Letters, 14, 947–952. Rotticci, D., Norin, T. and Hult, K. (2000) Organic Letters, 2, 1373–1376. Secundo, F. and Phillips, R.S. (1996) Enzyme and Microbial Technology, 19, 487–492. Terradas, F., Teston-Henry, M., Fitzpatrick, P.A. and Klibanov, A.M. (1993) Journal of the American Chemical Society, 115, 390–396. (a) Ke, T., Wescott, C.R. and Klibanov, A.M. (1996) Journal of the American Chemical Society, 118, 3366–3374. (b) Wescott, C.R., Noritomi, H. and Klibanov, A.M. (1996) Journal of the American Chemical Society, 118, 10365–10370. Colombo, G., Ottolina, G., Carrea, G., Bernardi, A. and Scolastico, C. (1998) Tetrahedron: Asymmetry, 9, 1205–1214. Saville, C.K. and Kazlauskas, R.J. (2005) Journal of the American Chemical Society, 127, 12228–12229. Hirose, Y., Kariya, K., Sasaki, J., Kurono, Y., Ebike, H. and Achiwa, K. (1992) Tetrahedron Letters, 33, 7157–7160. Nakamura, K., Takebe, J., Kitayama, T. and Ohno, A. (1991) Tetrahedron Letters, 32, 4941–4944. Ottosson, J., Fransson, L., King, J.W. and Hult, K. (2002) Biochimica et Biophysica Acta, 1594, 325–334. (a) Schmitke, J.L., Stern, L.J. and Klibanov, A.M. (1998) Proceedings of the National Academy of Sciences of the United States of America, 95, 12918–12923. (b) Schmitke, J.L., Stern, L.J. and Klibanov, A.M. (1998) Biochemical and Biophysical Research Communications, 248, 273–277. (c) Schmitke, J.L., Stern, L.J. and Klibanov, A.M. (1997) Proceedings of the National Academy of Sciences of the United States of America, 94, 4250–4255. Welton, T. (1999) Chemical Reviews, 99, 2071–2083.

j19

j 1 Medium Engineering

20

47 (a) van Rantwijk, F., Lau, R.M. and Sheldon, R.A. (2003) Trends in Biotechnology, 21, 131–138. (b) Park, S. and Kazlauskas, R.J. (2003) Current Opinion in Biotechnology, 14, 432–437. 48 Bonhote, P., Dias, A.P., Papageorgiu, N., Kalyanasundaram, K. and Gratzel, M. (1996) Inorganic Chemistry, 35, 1168–1178. 49 Kim, K.-W., Song, B., Choi, M.-Y. and Kim, M.-J. (2001) Organic Letters, 3, 1507–1509. 50 Schofer, S.H., Kafzik, N., Wasserscheid, P. and Kragl, U. (2001) Chemical Communications, 425–426. 51 Ulbert, O., Frater, T., Belaﬁ-Bako, K. and Gubicza, L. (2004) Journal of Molecular Catalysis B: Enzymatic, 31, 39–45. 52 Zhao, H. and Malhotra, S.V. (2002) Biotechnology Letters, 24, 1257–1260. 53 Zhao, H., Campbell, S.M., Jackson, L., Song, Z.Y. and Olubajo, O. (2006) Tetrahedron: Asymmetry, 17, 377–383. 54 Lou, W.Y., Zong, M.H., Wu, H., Xu, R. and Wang, J.F. (2005) Green Chemistry, 7, 500–506. 55 Piﬁssao, C. and Nascimento, M.P. (2006) Tetrahedron: Asymmetry, 17, 428–433.

56 Wallert, S., Drauz, K., Grayson, I., Groger, H., de Maria, P.D. and Bolm, C. (2005) Green Chemistry, 7, 602–605. 57 Quiros, M., Parker, M.-C. and Turner, N.J. (2001) The Journal of Organic Chemistry, 66, 5074–5079. 58 Okamoto, T., Yasuhito, E. and Ueji, S.-I. (2006) Organic and Biomolecular Chemistry, 4, 1147–1153. 59 Hsu, W.-T. and Clark, D.S. (2001) Biotechnology and Bioengineering, 73, 231–237. 60 Mine, Y., Fukunaga, K., Itoh, K., Yoshimoto, M., Nakao, K. and Sugimura, Y. (2003) Journal of Bioscience and Bioengineering, 95, 441–447. 61 Fasoli, E., Castillo, B., Santos, A., Silva, E., Ferrer, A., Rosario, E., Griebenow, K., Secundo, F. and Barletta, G.L. (2006) Journal of Molecular Catalysis B: Enzymatic, 42, 20–26. 62 Yu, J.-H. and Klibanov, A.M. (2006) Biotechnology Letters, 28, 555–558. 63 Ke, T. and Klibanov, A.M. (1999) Journal of the American Chemical Society, 121, 3334–3340.

j21

2 Directed Evolution as a Means to Engineer Enantioselective Enzymes Manfred T. Reetz

2.1 Introduction

As delineated in other chapters of this book, the world market for chiral organic compounds such as pharmaceuticals, plant protecting agents, and fragrances continues to expand, requiring efﬁcient methods for their synthesis [1,2]. Asymmetric catalysis constitutes the most efﬁcient and environmentally benign strategy to meet these synthetic goals. When choosing this approach, organic chemists have several options: transition metal catalysts [1,2], organocatalysts [3], and enzymes [4] or catalytic antibodies [5]. The decision as to which option is the best depends on a number of factors including cost of catalysts, catalyst stability, activity, and enantioselectivity, as well as type of solvent and ease of workup. A fundamentally new approach to the development of enantioselective catalysts for use in organic synthesis was introduced in the 1990s [6]. It makes use of directed evolution, a process that had previously been used to improve the stability and/or activity of enzymes [7]. The basis of this concept is the combination of proper molecular biological methods for random gene mutagenesis and expression coupled with high-throughput screening systems for the rapid identiﬁcation of enantioselective mutant enzymes. Whenever the natural (wild-type, WT) enzyme shows an unacceptably low enantioselectivity (ee) or selectivity factor (E value) for a given transformation of interest, A!B, a library of mutants is created from which the most highly enantioselective variant is identiﬁed. Then the process is repeated as often as necessary using, in each case, the gene (DNA segment) of an improved mutant for the next round of mutagenesis [6,8]. Since the inferior mutants are discarded, an evolutionary character of the overall process is simulated (Figure 2.1). In each cycle, the library of mutated genes is ﬁrst inserted in a standard bacterial host such as Escherichia coli or Bacillus subtilis. Subsequently, bacterial colonies are plated out on agar plates and harvested individually by a colony picker. Each colony is placed in a separate well of a microtiter plate containing nutrient broth, so that the bacteria grow and produce the protein of interest. Because each colony originates

Asymmetric Organic Synthesis with Enzymes. Edited by Vicente Gotor, Ignacio Alfonso, and Eduardo Garcıa-Urdiales Copyright 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 978-3-527-31825-4

j 2 Directed Evolution as a Means to Engineer Enantioselective Enzymes

22

Gene (DNA)

Wild-type enzyme

Random mutagenesis

1

2

3

4

5

...

5

... etc.

etc.

Library of mutated genes Repeat

Expression

1

2

3

4

Library of mutated enzymes Screening (or selection) for enantioselectivity 2

2

Positive mutants Figure 2.1 Strategy for directed evolution of an enantioselective enzyme [6,8].

from a single cell, mixtures of mutant enzymes are avoided. A portion of each mutant occurring in the harvested bacterial colony is then robotically placed on a different microtiter plate, where the enantioselective reaction of interest is carried out. Since the enzyme variants and the corresponding mutant genes are spatially addressable, the genotype/phenotype relation is maintained. The best mutant (positive hit) is then the starting point for the next cycle of gene mutagenesis, expression, and screening (Figure 2.2). Knowledge of the structure or mechanism of the enzyme is not required [6,8,9]. This approach is quite different from rational design based on site-directed mutagenesis. However, it is also logical owing to the evolutionary character of the process. As shown later in this chapter, the combination of rational design and randomization procedures constitutes an important option in directed evolution.

Mutagenesis

Repeat

Expression

Library of mutant genes in a test tube

.. . . . . .. . Colony . .. .. . .. .. . . . . . . picking . .. .

Bacterial colonies on agar plate

Screening for enantioselectivity Bacteria producing mutant enzymes grown in nutrient broth

Visualization of positive mutants

Figure 2.2 The experimental stages of directed evolution of enantioselective enzymes [6,8].

2.2 Molecular Biological Methods for Mutagenesis

Following a number of early contributions, the 1980s witnessed signiﬁcant progress in the development of efﬁcient methods for gene mutagenesis and expression, and libraries of mutant enzymes were constructed and screened for enhanced stability or activity [10]. However, this process in itself is not yet directed evolution, because only one cycle of mutagenesis/screening is involved. Rare cases of at least two such cycles began to appear in the literature, which constitute the ﬁrst cases of true directed evolution of enzymes [10c]. However, it was not until the period of 1993–1997 that the general idea of directed evolution was fully appreciated [11]. Using known and new methods of gene mutagenesis in combination with screening or selection systems, enzyme properties such as activity, thermostability, and stability against hostile solvents were improved. These developments set the stage for the concept of directed evolution of enantioselective enzymes early on (Figures 2.1 and 2.2) [6,8]. The challenges in this new ﬁeld of asymmetric catalysis include the elaboration of optimal strategies for scanning protein sequence space for enantioselectivity and the development of efﬁcient high-throughput screening systems for evaluating thousands of potentially enantioselective mutant enzymes.

2.2 Molecular Biological Methods for Mutagenesis

During the last two decades, numerous gene mutagenesis methods have become available for application in directed evolution [7,11e,12]. Some of the most important gene mutagenesis methods are described brieﬂy here. For complete coverage, the reader is referred to recent reviews [11e,12,13]. The use of mutator strains or the treatment of the isolated DNA with UV light or chemicals, as reported in the older literature and sometimes still used today, provides a means to target the whole gene for more or less random point mutagenesis, but it is not trivial to control such processes [7a]. Modern gene mutagenesis methods can be roughly assigned to two different categories: those based on point mutations and those making use of recombination [7,11e,12]. The most frequently used mutagenesis method in directed evolution is error-prone polymerase chain reaction (epPCR), described by Leung [14] and improved by Cadwell and Joyce [15]. It is based on the classical DNA ampliﬁcation method, polymerase chain reaction (PCR). By varying parameters such as the concentration of MgCl2 (or MnCl2) and/or nucleotide concentration, errors in DNA ampliﬁcation are induced. The mutation rate can be adjusted empirically so that on average one, two, or more amino acid exchange events occur in the expressed protein. Although epPCR is sometimes considered to induce point mutations randomly over the entire gene (and protein), this is actually not the case. Owing to the degeneracy of the genetic code, inter alia, some amino acid exchanges are favored in a given enzyme whereas others are less so [7,16]. Nevertheless, most directed evolution projects begin with epPCR because it is very easy to perform and provides a convenient platform from which the actual evolutionary optimization can begin, very often by additional cycles of epPCR.

j23

j 2 Directed Evolution as a Means to Engineer Enantioselective Enzymes

24

The major challenge in putting directed evolution into practice has to do with the problem of probing protein sequence space efﬁciently, so that the screening effort can be minimized. The number of enzyme variants N at the theoretically maximum degree of diversity is given by the following algorithm: N¼

19M X! ðX MÞ!M!

ð2:1Þ

where M is the total number of amino acid exchanges per enzyme molecule and X is the number of amino acids per enzyme molecule [7]. For the purpose of illustration, an enzyme composed of 300 amino acids theoretically has 5700 mutants if one amino acid is exchanged, but 16 million in the case of two simultaneous exchanges, and 30.5 billion if three amino acids are exchanged per enzyme molecule simultaneously [7]. Such numbers are often cited to illustrate the problem of protein sequence space, which of course points to the difﬁculty in screening such large numbers of enzyme mutants. The formation of focused mutant libraries also provides the experimenter with a useful tool in directed evolution [7]. To apply such methods, information regarding the structure of the enzyme (X-ray data or homology model), may be necessary. An intelligent decision is then made regarding the identiﬁcation of a hot spot, a position in the amino acid sequence of the enzyme that is believed to be important with respect to a given catalytic property, for example, near the binding pocket. Following this choice, the position is randomized by saturation mutagenesis leading to an enzyme library containing the 20 theoretically different mutants corresponding to the 20 proteinogenic amino acids at a deﬁned position [7,17]. To ensure >95% coverage (i.e. to make certain that all 20 mutants have been evaluated in an appropriate screen), oversampling is necessary [18]. (For a computer program (CASTER) useful in designing saturation mutagenesis libraries, see homepage of the author: www.mpi-muelheim.mpg.de/ reetz.html). It usually sufﬁces to screen about 100–200 clones for a single amino acid exchange(i.e.topickand evaluate thisnumber of bacterialcolonies ontheagarplates).In a different approach, knowledge of the 3D structure of the enzyme is not required because every single position is randomized separately in a systematic manner [19,20]. In still another strategy, hot spots resulting from epPCR are each randomized [21]. In an important extension of the above approaches, cassette mutagenesis was developed, which is a form of saturation mutagenesis [11e,17,22]. The user identiﬁes (or predicts) a hot region in the enzyme, either on the basis of structural and mechanistic data or as a result of applying epPCR. One usually focuses on a portion of the enzyme near the binding site and designs a cassette, which is a DNA fragment consisting of a deﬁned number of nucleotides encoding the desired amino acids. If two amino acid positions are chosen for simultaneous randomization, 400 theoretically different mutants are possible, and the necessary oversampling for 95% coverage of protein sequence space can be handled using current screening methods (3000 clones). However, when the number of amino acid positions for randomization is increased, screening problems arise if complete coverage is strived for. For example, if four positions are randomized simultaneously, catalyst diversity is high (204 ¼ 160 000 different mutants), which may be desirable, but oversampling necessary for 95% coverage requires >3 million clones to be screened. Of course, one can

2.2 Molecular Biological Methods for Mutagenesis

j25

settle for much lower coverage, and the greatly reduced catalyst diversity in the screened mutants can still lead to the discovery of hits displaying improved catalytic proﬁles, as has been demonstrated, for example, in the directed evolution of enantioselective lipases [22]. Recently, a method called iterative saturation mutagenesis (ISM) has been proposed and experimentally implemented, which appears to be exceptionally efﬁcient [23,24]. ISM reduces the screening effort drastically, which means that it can be considered to constitute accelerated directed evolution. It is based on iterative cycles of saturation mutagenesis at predetermined sites in an enzyme, a given site being composed of one or more amino acid positions. The basis for choosing these sites is crucial and depends upon the nature of the catalytic property to be improved. In the case of substrate acceptance and/or enantioselectivity, the choice is made using the combinatorial active-site saturation test (CAST) [23,25–27]. Accordingly, an X-ray structure or a homology model is used to identify appropriate sites around the complete binding pocket. In the case of thermostability, sites showing highest B-factors, likewise available from X-ray data, are chosen [28]. The most general scheme of ISM is given in Figure 2.3, illustrating the case involving four sites. In a simpliﬁed version of ISM, each site is considered only once (Figure 2.4). This means that, in the given case of four sites, 65 libraries of mutants are possible, the overall evolutionary process being convergent. Of course, not all branches have to be explored. Indeed, this embodiment of ISM has been shown to be highly successful in the rapid evolution of enantioselectivity [23,26,27] and thermostability [28]. The reason(s) for the success of ISM has to do with the fact that sequence space has been conﬁned to deﬁned locations in the protein that are most likely to respond positively in an additive or cooperative manner. In sharp contrast, when performing several rounds of epPCR (4–8 are typical!), the whole protein is addressed repeatedly although only a fraction of the amino acid positions are important. Owing to statistical reasons, a given improved mutant (hit) evolved by ISM is not etc.

A CD

B

ABD

C

ABC

BCD

D

A BD

A

A

C

A BC

D

BCD

B

A

B

A BC

A CD

A

D

C

WT Figure 2.3 Schematic illustration of Iterative Saturation Mutagenesis (ISM) involving (as an example) four randomization sites A, B, C and D [23,24].

ACD

B CD

D

B

ABD

C

C

D

D

B B

C C

B C

D

A

D

A

A

C D

C

D

D

A

B

A

C C

A B

D

A

C

D A

B D

B

D

D

A A

B B

D

A B

C

A

B

C

C

A C

A

B C

WT Figure 2.4 ISM employing four sites A, B, C and D, each site in a given upward pathway being visited only once.

B

D

C

B

C D

D

A

B

B

A

26

j 2 Directed Evolution as a Means to Engineer Enantioselective Enzymes

2.3 High-throughput Screening Methods for Enantioselectivity

Figure 2.5 Scheme for DNA shuffling illustrated for the case in which the parent genes originate from the WT by some form of mutagenesis [7e].

likely to be found by applying the traditional strategy based on repeating cycles of epPCR. Rather than point mutations, recombination can be considered, which in a genetic sense means the breaking and rejoining of DNA in new combinations. The most prominent version, pioneered by Stemmer in 1994, is DNA shufﬂing [7e,11c,11d]. In general, one or more genes are ﬁrst digested with a DNase to yield double-stranded oligonucleotide fragments of 10–50 base pairs, which are then ampliﬁed in a PCRlike process. Repeating cycles of strand separation and reannealing in the presence of a DNA polymerase followed by a ﬁnal PCR-ampliﬁcation step results in the reassembly of full-length mutant genes. DNA shufﬂing can be performed with one gene, with two or more natural genes, or with mutant genes (Figure 2.5). A limitation is the requirement of relatively high homology. If homologous genes from different species are chosen, the process is called family shufﬂing, which appears to be particularly efﬁcient because it provides high catalyst diversity [7e,29]. A certain degree of self-hybridization of parental genes occurs, which lowers the quality of the mutant libraries (some colonies contain WT). Other recombinant methods have also been developed, each having advantages and (some) disadvantages [7,12]

2.3 High-throughput Screening Methods for Enantioselectivity

A number of high-throughput enantiomeric excess assays have been developed, yet none are completely general. This crucial aspect of directed evolution of

j27

j 2 Directed Evolution as a Means to Engineer Enantioselective Enzymes

28

enantioselective enzymes is not treated here and the reader is referred to several reviews [30]. Sufﬁce it to say that many of the screens allow typically 1000–5000 samples to be evaluated per day. If this is not possible in a given case, mediumthroughput GC [31] or HPLC [32] should be used, meaning typically 300–800 enantiomeric excess determinations per day. A prescreen for activity, which is easier than enantiomeric excess determination, is advisable because this eliminates nonactive clones.

2.4 Examples of Enhancing the Enantioselectivity of Enzymes by Directed Evolution

Inspite of the availability of numerous mutagenesis methods and enantiomeric excess assays, the best strategy for performing directed evolution of enantioselectivity of a given enzyme for an asymmetric reaction has not yet been established. Following the ﬁrst report of directed evolution of an enantioselective enzyme [6], which focused on a lipase, numerous individual studies regarding further successful examples have followed [33]. However, only one system, namely, the original lipase-catalyzed reaction, has been studied systematically with respect to illuminating the relative merits of applying the various mutagenesis methods. Therefore, the results and conclusions of these ongoing studies are presented here in detail. 2.4.1 Lipase from Pseudomonas aeruginosa (PAL)

The ﬁrst case study of the directed evolution of an enantioselective enzyme involving repeating cycles of gene mutagenesis, expression, and screening (Figures 2.1 and 2.2) was carried out with Pseudomonas aeruginosa lipase (PAL) [6]. Lipases (triacylglycerol ester hydrolases, EC 3.1.1.3) [4,34] are the most widely used enzymes in synthetic organic chemistry, catalyzing the chemo-, regio- and/or stereoselective hydrolysis of carboxylic acid esters or the reverse reaction in organic solvents. In the case of enantioselectivity, either desymmetrization of prochiral substrates or kinetic resolution of racemates is relevant, and numerous academic and industrial examples are known [34]. The active site of lipases is characterized by a triad composed of serine, histidine, and aspartate, the oxyanion being the crucial intermediate [34] (Figure 2.6). The kinetic resolution of rac-1 was chosen as a model reaction using the WT lipase from PAL as the catalyst [6]. The WT shows a very low selectivity factor: E ¼ 1.1 in slight favor of (S)-2 (Scheme 2.1). PAL is composed of 285 amino acids, the active site being Ser82 [35]. Initially (in 1995–1996), epPCR at a low mutation rate was chosen to induce an average of only one amino acid exchange per enzyme molecule [6]. This choice was in line with the general opinion of other researchers in the area of directed evolution at that time, although their goals were different (enhanced stability, etc.). Typically, 2000–3000 enzyme variants per generation were screened using a UV–vis–based screening

2.4 Examples of Enhancing the Enantioselectivity of Enzymes by Directed Evolution

H N

N

H O

Catalytic triad Ser

O O

O

O

OR2

O

ON H

O Substrate

Ser

R1

OR2

R1

O-

O

Ser

His

Asp

j29

+

R2OH

R1

H H Acyl enzyme intermediate

‘Oxyanion’

Aroduct alcohol

Ser + O H

R1CO2H Product acid

Figure 2.6 Mechanism of lipase-catalyzed hydrolysis of esters [34].

O

O

R O

NO2

CH3 rac-1 (R = n-C8H17)

H2O Lipase-variants

R

O OH +

CH3 (S)-2

R

OH + O

NO2

CH3 (R)-2

3

Scheme 2.1

system. The (R) substrate and (S) substrate were studied separately pairwise to assess enantioselectivity. As a result of four cycles of mutagenesis/screening, a variant (I) having an E value of 11.3 in favor of (S)-2 was identiﬁed (Figure 2.7). Although this constitutes proof-of-principle, a factor of E ¼ 11.3 is far from practical. A ﬁfth round of epPCR was shown to increase the E value slightly, but such a strategy is far from optimal. Indeed, it became clear that it is better to apply only two or three rounds of epPCR and then to switch to another mutagenesis method [8,36]. One of these methods is saturation mutagenesis. It was assumed that the observed sites of amino acid exchange produced by epPCR are hot spots that are important for increasing enantioselectivity, but that the speciﬁc amino acids identiﬁed by sequencing are not necessarily optimal [21c]. Therefore, several of these positions were chosen for saturation experiments [11e,21e,37]. The same strategy was developed independently in a study concerning the enhancement of the thermostability of a protease [21e,38], and it is now used routinely in directed evolution studies. However, not all of the original hot spots constitute sites that lead to improvements upon applying saturation mutagenesis. One of the successful positions turned out to be 155 (Figure 2.7), which was randomized using the appropriate oligonucleotides. This strategy led to variant II displaying an E value of 20 [21c]. The gene encoding enzyme

j 2 Directed Evolution as a Means to Engineer Enantioselective Enzymes

30

E = 11.3 E = 9.4

E = 4.4 E

S155L S149G

V476 S155L S149G

Enzyme variant I F259L V476 S155L S149G

E = 2.1 S149G

E = 1.1 0

WT 1 2 3 Mutant generations

4

Figure 2.7 Enhanced E values of the PAL-catalyzed hydrolysis of rac-1 by cumulative mutations caused by four rounds of epPCR [6].

variant II was then subjected to another cycle of epPCR, resulting in an even better enzyme variant III having ﬁve mutations and a selectivity factor of E ¼ 25 in favor of the (S) substrate. In order to assess the utility of other mutagenesis methods, DNA shufﬂing was considered, but it was not clear which mutant genes should be shufﬂed. DNA shufﬂing of the mutant genes produced by low-mutation-rate epPCR (Figure 2.7) failed to achieve signiﬁcant improvements, leading to the conclusion that higher gene diversity is necessary [22]. Therefore, epPCR of the WT gene was repeated, this time at a higher mutation rate corresponding to an average of three amino acid exchange events per enzyme molecule. This process led to the identiﬁcation of several improved enzyme variants, IV and V (E ¼ 3 and E ¼ 6.5) being the best ones [22]. In the original project using a low mutation rate, such selectivity factors were not obtained until the second cycle of epPCR (Figure 2.7) [6]. Thus, it appears that high mutation rates are more successful. The genes encoding the variants III, IV, and V were then subjected to conventional DNA shufﬂing. This process provided variant VI having the highest enantioselectivity up to that point (E ¼ 32) [22]. However, in a different strategy, a focused library was generated by randomizing amino acids 160–163, a region next to the binding site (Figure 2.8) [22]. This procedure constitutes a fusion of rational design and directed evolution. For this purpose saturation mutagenesis was applied, resulting in simultaneous randomization at all four positions. This provided variant VII with E ¼ 30, which is characterized by four mutations (E160A, S161D, L162G, and N163F) [22]. In this case only 5000 clones were screened, although for 95% coverage of the 160 000 theoretically different mutants an oversampling of about 3 million clones would have been necessary. Clearly, all of these strategies for navigating in protein sequence space are successful, but it is not obvious which one is optimal. Finally, the region of accessible protein sequence space was extended by developing a modiﬁed version of Stemmers combinatorial multiple-cassette mutagenesis (CMCM)

2.4 Examples of Enhancing the Enantioselectivity of Enzymes by Directed Evolution

Figure 2.8 Binding pocket of PAL for the acid part of rac-1 showing the geometric position of amino acids 160–163, which were randomized simultaneously by saturation mutagenesis [22].

[22]. The original form of CMCM is a special type of DNA shufﬂing using the WTgene and cassettes composed of deﬁned sequences to be randomized [39]. In the lipase project, two mutant genes encoding the enzyme variants IV and V and a mutagenic oligocassette allowing simultaneous randomization at the hot spots 155 and 162 were subjected to DNA shufﬂing. This resulted in the most enantioselective variant X, displaying a selectivity factor of E ¼ 51 [22]. It is characterized by six mutations, most of which are remote from the active site (Ser 82) (see the discussion that follows). The results of these and other experiments are summarized in Figure 2.9. A total of 40 000 mutants were screened. It is likely that, upon exploring larger portions of protein sequence space, even better lipase variants can be identiﬁed. However, efﬁcient directed evolution is not a matter of generating huge libraries, which then require considerable efforts in screening; the goal is to rather create high-quality libraries while minimizing their size [7,8c,12]. These initial systematic studies regarding the directed evolution of PAL allowed several conclusions to be made. Protein sequence space can be explored successfully by applying the following strategies [8c,33]: (1) generation of mutants by several cycles of epPCR at low or preferably at high mutation rate; (2) identiﬁcation of hot spots and application of saturation mutagenesis; (3) identiﬁcation of hot regions empirically or by rational design and subsequent application of saturation mutagenesis; (4) application of epPCR at high mutation rate followed by DNA shufﬂing of the mutants; (5) extension of the process of CMCM to cover focused positions in the enzyme (hot spots identiﬁed by earlier experiments).

j31

4 cycles of epPCR at low mutation rate

Enzyme variant I E = 11

Saturation mutagenesis at hot spots

Enzyme variant II E = 20

Wild-type E = 1.1

1 cycle of epPCR at high mutation rate

Enzyme variants IV with E = 3 V with E = 6.5

epPCR at high mutation rate

No significant improvement

Figure 2.9 Schematic summary of the directed evolution of enantioselective lipase variants originating from the WT PAL used as catalysts in the hydrolytic kinetic resolution of ester rac-1. CMCM ¼ Combinatorial multiple-cassette mutagenesis [8c,22].

Further epPCR at low mutation rate

Small improvements

epPCR at low mutation rate

Enzyme variant III E = 25

epPCR at low or high mutation rate

No significant improvement

Modified CMCM with IV, V and oligo-cassette at positions 155/162

Enzyme variant X E = 51

DNA-shuffling with III, IV and V

Enzyme variant VI E = 32

: Generated variant

: Mutagenesis method

Cassette mutagenesis region 160-163

Enzyme variant VII E = 30

Cassette mutagenesis at positions 155/162

Enzyme variants VIII with E = 34 IX with E = 30

32

j 2 Directed Evolution as a Means to Engineer Enantioselective Enzymes

2.4 Examples of Enhancing the Enantioselectivity of Enzymes by Directed Evolution

Efforts were also made to invert the sense of enantioselectivity in the hydrolytic kinetic resolution of ester (1) using PAL with preferential formation of (R)-2 [40,41]. Using epPCR and DNA shufﬂing, an (R)-selective mutant showing an E value of 30 was evolved by screening about 45 000 clones for the (R) enantiomer. The best mutant is characterized by 11 mutations, which are different from those of the best (S)-selective variant X [41]. Systematic investigations of the type described above provide two kinds of lessons. First, they offer some hints on how to choose the right strategy for probing protein sequence space [8c]. Second, they provide intriguing data for theoretical studies concerning the origin of enantioselectivity [36]. Most of the above results were obtained without prior knowledge of the 3D structure of the enzyme, or of its mechanism; rather, the Darwinian nature of the concept was relied on. Insight into how enzymes function can be gained from directed evolution provided a sound theoretical analysis is performed. This in turn generates information, which can be used to simplify further genetic optimization. The lipase (PAL) used in these studies is a hydrolase having the usual catalytic triad composed of aspartate, histidine, and serine [42] (Figure 2.6). Stereoselectivity is determined in the ﬁrst step, which involves the formation of the oxyanion. Unfortunately, X-ray structural characterization of the (S)- and (R)-selective mutants are not available. However, consideration of the crystal structure of the WT lipase [42] is in itself illuminating. Surprisingly, it turned out that many of the mutants have amino acid exchanges remote from the active site [8,22,40]. In order to approximate the activity of the best variant X, the supernatants were used to perform kinetics [8c]. The kcat values of variant X turned out to be 9.6 s1 for (S)-1 and 1.2 s1 for (R)-1. In going from the WT to variant X, the kcat/Km value (l mol1 s1) increased signiﬁcantly: for (S)-1, kcat/Km ¼ 9.0 · 102 (wild type) and 3.7 · 105 (variant X); for (R)-1, kcat/Km ¼ 3.5 · 102 (wild type) and 8.4 · 103 (variant X). Thus, although these are only rough numbers, they show that variant X is much more active than the WT. In order to study the source of enhanced (S) selectivity, variant X and other real and hypothetical mutants were subjected to a detailed molecular mechanics/quantum mechanical (MM/QM) study [36]. The six mutations (amino acid substitutions) of variant X are D20N, S53P, S155M, L162G, T180I, and T234S. With the exception of L162G, they all occur at positions remote from the active center (S82). Remote effects had been observed in earlier investigations of traditional protein engineering [10a], and directed evolution [7,43,44], but these were all cases in which thermostability, chemical stability, or activity was the focus of interest. In contrast, enantioselectivity is a parameter that chemists, biochemists, and enzymologists traditionally associate with the active center and the binding pocket, in the tradition of Emil Fischers hypothesis of lock-and-key or Koshlands model of induced ﬁt [45]. Indeed, all previous attempts to inﬂuence the enantioselectivity of an enzyme based on site-speciﬁc mutagenesis as suggested by rational design focused on the binding pocket [9]. The initial results of the MM/QM study regarding the source of enhanced enantioselectivity led to several plausible conclusions [36a]. First, only two of the six amino acid substitutions of mutant X inﬂuence enantioselectivity substantially.

j33

j 2 Directed Evolution as a Means to Engineer Enantioselective Enzymes

34

Figure 2.10 The oxyanions originating from rac-1 in the WT PAL (left) and mutant X (right) [8c,36].

Second, there is essentially no difference in the binding energetics of the (S) and (R) substrates; rather, stereodifferentiation occurs as the oxyanion is formed. Third, a novel relay mechanism starting at the surface of the enzyme and working inward toward the active site is operating. Accordingly, the H-bond network deﬁned by S161/ S53/H83 in the WT is disrupted by the mutation S53P, setting the sidechain of H83 free to move (Figure 2.10). At the same time, mutation L162G provides additional space for the substrate (larger binding pocket) and also for the imidazole residue of H83 to move toward the active site. In the case of the preferred enantiomer (S)-1, additional stabilization of the oxyanion via H-bonding occurs. This is in contrast to (R)-1, in which case the corresponding oxyanion cannot proﬁt from such stabilization for steric reasons. It is likely that the same effects operate in the transition state of the reaction leading to the oxyanion [36]. These results are fully in line with the conclusions of the observed higher activity of variant X relative to the WT [8b]. Lessons continued to be learnt from directed evolution as a result of careful theoretical analysis in a follow-up theoretical study [36b]. For example, it was predicted that a double mutant characterized by S53P/L162G should also be an excellent catalyst. Therefore, this mutant as well as other mutational combinations resulting from deconvolution of the previously most enantioselective variant having six mutations were prepared and tested. It turned out that the S53P/L162G mutant is even more enantioselective (E ¼ 64) [36b]. In addition to clearing up the reasons for the enhancement of enantioselectivity, this study also raises important questions regarding the efﬁciency of the various strategies used in probing protein sequence space. Apparently, more than half of the six mutations in the earlier best mutant X (E ¼ 51) are superﬂuous, which may have several reasons. Systematic experimental and theoretical studies of this kind are helpful in performing in vitro evolution of enantioselectivity. Nevertheless, several questions are not fully answered. Are remote mutations more important than those close to the active site, or is the opposite true? Is it more effective to allow randomization all over the enzyme rather than focusing on the region around the active site (or vice versa)? To be sure, when applying epPCR or any other mutagenesis method that more or less

2.4 Examples of Enhancing the Enantioselectivity of Enzymes by Directed Evolution

covers the whole gene, there are statistically more remote amino acids for exchange events to occur than amino acids close to the active site. Application of the recently proposed ISM (see Section 2) [23,24,26–28] using the combinatorial active-site saturation test (CAST) [25] may turn out to be the most efﬁcient way to evolve enantioselectivity rapidly. CASTing was initially applied in order to expand the range of substrate acceptance, speciﬁcally of PAL. The ﬁrst step in CASTing is to inspect the X-ray structure or a homology model and to identify sites (A, B, C, D, etc.) at which the amino acid sidechains reside next to the binding pocket. Each site may be composed of one, two or more amino acid positions (usually not more than three). The sites are then subjected to saturation mutagenesis using appropriate nucleotide cassettes with formation of highly focused libraries, which are then screened for activity and/or enantioselectivity. Thus, CASTing means the generation of focused libraries around the complete binding pocket [25]. This distinguishes it from previous studies of focused libraries in which only speciﬁc sites were selected [7], as for example, in the simultaneous randomization of amino acids 160–163 of PAL [22]. In the case of CASTing for enhanced substrate acceptance of PAL, inspection of the X-ray structure [42] led to the identiﬁcation of ﬁve sites, A, B, C, D, and E, around the binding pocket harboring the acid moiety of esters [25]. The enzyme was known to catalyze the hydrolysis of fatty acid esters, but not of bulky esters. The initial study focused not only on the original ester (1) but also on the sterically more demanding substrates (4–13). Most of these p-nitrophenyl esters are not accepted by the WT (some background reaction is observed owing to the fact that they are activated esters). OR OR

O

O OR

O 1

4

5

O

OR O

OR 6

OR O

7

8

O

O 9

O

O

H3CO

10

11

O R = p-nitrophenyl

OR OR 12

OR

OR

OR

13

j35

j 2 Directed Evolution as a Means to Engineer Enantioselective Enzymes

36

Figure 2.11 CASTing of the lipase from Pseudomonas aeruginosa (PAL) leading to the construction of five libraries of mutants (A–E) produced by simultaneous randomization at sites composed of two amino acids. (For illustrative purposes, the binding of substrate (1) is shown) [25].

The following amino acid pairs were deﬁned: M16/L17, L118/I121, L131/V135, L159/L162 and L231/V232. Five libraries A, B, C, D, and E, respectively, were then created separately by simultaneous randomization at each pair (Figure 2.11) [25]. All the ﬁve libraries of mutants were then tested in the hydrolysis of substrates (1) and (4–13). It turned out that most of the hits originate from libraries A or D. Figure 2.12 summarizes the relative rates of the WT- and mutant-catalyzed hydrolyses, the eight best mutants being featured. The data in Figure 2.12 are the results of initial mutagenesis experiments, which does not yet constitute directed evolution. An evolutionary process was subsequently induced by combining the mutations of two improved mutants of the ﬁrst round [46]. Thereby new mutants were obtained, which show an increase of activity relative to the WT by more than 2 orders of magnitude. Although enantioselectivity was not the

2.4 Examples of Enhancing the Enantioselectivity of Enzymes by Directed Evolution

Figure 2.12 Substrate profiles of lipase variants produced by CASTing [25].

j37

j 2 Directed Evolution as a Means to Engineer Enantioselective Enzymes

38

focus of this study, some of the mutants were examined for this property. For example, one of the evolved mutants showed a selectivity factor of E ¼ 49 in the hydrolytic kinetic resolution of the bulky substrate (8) [46]. ISM needs to be tested for comparison. 2.4.2 Other Lipases

The Bacillus subtilis lipase A (BSLA) was the subject of two short directed evolution studies [19,47]. In one case systematic saturation mutagenesis at all of the 181positions of BSLA was performed [19]. Using meso-1,4-diacetoxy-2-cyclopentene as the substrate, reversed enantioselectivity of up to 83% ee was observed. In another study synthetic shufﬂing (Assembly of Designed Oligonucleotides) was tested using BSLA [47]. In yet another investigation, inversion of enantioselectivity was accomplished by a different approach [48]. The focus of this study was the Burkholderia cepacia KWI-56 lipase as a catalyst in the hydrolytic kinetic resolution of rac-14 (Scheme 2.2). The WT shows (S) selectivity with an E factor of 33. Using X-ray structural data of the lipase, a decision was made concerning the choice of four amino acid residues near the binding product. This site was then randomized using seven hydrophobic amino acids building blocks. By use of a novel in vitro technique for the construction and screening of a protein library by single-molecule DNA ampliﬁcation via PCR followed by in vitro coupled transcription/translation, a mutant lipase was identiﬁed showing reversed enantioselectivity (E ¼ 38) [48]. It would be of interest to test iterative CASTing and to compare the results. Ph

Ph CO2Et

rac-14

Ph CO2H

(R)-15

+

CO2H (S)-15

Scheme 2.2

2.4.3 Esterases

Esterases have a catalytic function and mechanism similar to those of lipases, but some structural aspects and the nature of substrates differ [4]. One can expect that the lessons learned from the directed evolution of lipases also apply to esterases. However, few efforts have been made in the directed evolution of enantioselective esterases, although previous work by Arnold had shown that the activity of esterases as catalysts in the hydrolysis of achiral esters can be enhanced [49]. An example regarding enantioselectivity involves the hydrolytic kinetic resolution of racemic esters catalyzed by Pseudomonas ﬂuorescens esterase (PFE) [50]. Using a mutator strain and by screening very small libraries, low improvement in enantioselectivity was

2.4 Examples of Enhancing the Enantioselectivity of Enzymes by Directed Evolution

observed [50]. Moderate effects were observed upon applying saturation mutagenesis at hot spots [51]. It was concluded that distal mutations increase enantioselectivity only moderately and that mutations near the active site exert stronger effects [51a]. However, in a later study using the same enzyme PFE and applying epPCR, it was postulated that remote mutations are more important [52]. In another study a hyperthermophilic esterase from Aeropyrum pernix K1 (APE1547) was used as a catalyst in the hydrolytic kinetic resolution of rac-3-octanol acetate [53]. Following a single round of epPCR, a mutant displaying a 2.6-fold increase in enantioselectivity was identiﬁed having ﬁve amino acid substitutions, which were shown to be spatially distal to the catalytic center. 2.4.4 Hydantoinases

Hydantoinases belong to the E.C.3.5.2 group of cyclic amidases, which catalyze the hydrolysis of hydantoins [4,54]. As synthetic hydantoins are readily accessible by a variety of chemical syntheses, including Strecker reactions, enantioselective hydantoinase-catalyzed hydrolysis offers an attractive and general route to chiral amino acid derivatives. Moreover, hydantoins are easily racemized chemically or enzymatically by appropriate racemases, so that dynamic kinetic resolution with potential 100% conversion and complete enantioselectivity is theoretically possible. Indeed, a number of such cases using WT hydantoinases have been reported [54]. However,ifasymmetricinductionispoororifinversionofenantioselectivityisdesired, directed evolution can come to the rescue. Such a case has been reported, speciﬁcally in the production of L-methionine in a whole-cell system (E. coli) (Figure 2.13) [55]. Following epPCR and saturation mutagenesis at hot spots, the D-selective hydantoinase from Arthrobacter sp. DSM 9771 was converted into L-selective variants. The best L-selective mutant showed a value of 20% ee at about 30% conversion, compared to the WT displaying ee ¼ 40% in favor of the D-methionine-derivative. With the help of an appropriate L-carbamoylase, L-methionine itself was produced. This academic/ industrial effort provided several selective hydantoinases of industrial interest (O May, (Degussa-H€ uls), personal communication, 2005). 2.4.5 Nitrilases

Nitrilases catalyze the synthetically important hydrolysis of nitriles with formation of the corresponding carboxylic acids [4]. Scientists at Diversa expanded the collection of nitrilases by metagenome panning [56]. Nevertheless, in numerous cases the usual limitations of enzyme catalysis become visible, including poor or only moderate enantioselectivity, limited activity (substrate acceptance), and/or product inhibition. Diversa also reported the ﬁrst example of the directed evolution of an enantioselective nitrilase [20]. An additional limitation had to be overcome, which is sometimes ignored, when enzymes are used as catalysts in synthetic organic chemistry: product inhibition and/or decreased enantioselectivity at high substrate concentrations [20].

j39

j 2 Directed Evolution as a Means to Engineer Enantioselective Enzymes

40

L-enantiomer

RH 5-Monosubstituted hydantoin

HN

D-enantiomer

O

Racemase

NH

or pH > 8

HR HN

Hydantoinase

+ H2O O

+ H2O

H R

O OH NH2

OH NH2

N-carbamoyl-amino acid

O

O

RH H2N

X

L-carbamoylase

+ H2O

α-amino acid

NH O

O

R H

O

O

+ CO2

OH

+ NH3

Figure 2.13 Reactions and enzymes involved in the production of L-amino acids from racemic hydantoins by the three-enzyme hydantoinase process [55].

The study concerns the desymmetrization of the prochiral dinitrile (16) with preferential formation of the (R)-17, which was known to be a chiral intermediate in the synthesis of the cholesterol-lowering therapeutic drug (18) (Lipitor, Sortis, Torvast, etc.) as shown in Scheme 2.3. Upon screening genomic libraries obtained from environmental samples, more than 200 new nitrilases that allow mild and selective hydrolysis of nitriles were discovered [56]. One of them catalyzes the (R)-selective hydrolysis of (16) with a value O-

HO

O OH OH NC

CN Nitrilase 16

OH

H2O NC

Several CO2H

F

N

steps

CH3

(R)-17

O HN 18

Scheme 2.3

CH3

2.4 Examples of Enhancing the Enantioselectivity of Enzymes by Directed Evolution

of 94.5% ee at a substrate concentration of 100 mM [20]. However, when experiments were done at a more practical concentration of 2.25 M, activity and enantioselectivity suffered (only 87.8% ee). Therefore, directed evolution was applied to solve these problems. Saturation mutagenesis was performed systematically at all positions of the 330-amino acid enzyme [20]. The library contained 10 528 genetic variants and was screened using a mass spectrometry-based high-throughput enantiomeric excess assay. Of the 31 584 clones that were considered, 17 led to enhanced enantioselectivity over the WT enzyme. Each of these variants was then tested at 2.25 M substrate concentration. Many performed poorly at this high concentration, but several mutants having newly introduced serine, histidine, or threonine at position 190 (hot spot!) showed enhanced enantioselectivity. The A190H mutant turned out to be the most enantioselective and active catalyst. Within 15 hours, complete conversion of (16) was observed, affording (R)-17 with 98% ee. This dramatic improvement is contrasted with the performance of the less active WT (88% ee). The system works even at 3 M substrate concentration with 90% conversion, 98.5% ee, and essentially no substrate inhibition. Volumetric productivity, which is of great importance in any industrial application, was enhanced by the process of directed evolution [20]. 2.4.6 Epoxide Hydrolases

Several reports regarding the directed evolution of enantioselective epoxide hydrolases (EHs) have appeared [23,57–59]. These enzymes constitute important catalysts in synthetic organic chemistry [4,60]. The ﬁrst two reported studies concern the Aspergillus niger epoxide hydrolase (ANEH) [57,58]. Initial attempts were made to enhance the enantioselectivity of the ANEH-catalyzed hydrolytic kinetic resolution of glycidyl phenyl ether (rac-19). The WT leads to an E value of only 4.6 in favor of (S)-20 (see Scheme 2.4) [58]. Several libraries of mutant ANEHs were prepared by applying epPCR at various mutation rates and transforming into E. coli BL21 (DE3). About 20 000 clones were screened, the most selective ANEH variant showing a selectivity factor of only E ¼ 10.8 in the kinetic resolution of rac-19 [58]. Thus, this enzyme appeared to be difﬁcult to evolve. The ANEH-mutant displaying enhanced enantioselectivity (E ¼ 10.8) was sequenced and shown to be characterized by three mutations, A217V near the active site and K332E and A390E both at remote positions [58]. The X-ray crystal structure of the WT ANEH had been analyzed earlier [61], revealing a dimer comprising identical

O PhO rac-19 Scheme 2.4

HO

H 2O EH from A. niger

HO

OH + PhO

PhO (S)-20

(R)-20

OH

j41

j 2 Directed Evolution as a Means to Engineer Enantioselective Enzymes

42

subunits. Each monomeric unit contains tyrosines, which bind and activate the epoxide and the catalytic triad (two aspartates and a histidine), typical of most EHs. Aspartate 192 attacks the epoxide nucleophilically with formation of a glycolmonoester intermediate that is hydrolyzed in a second step. The binding pocket is composed of a very narrow tunnel, which may explain the difﬁculty in evolving enhanced enantioselectivity. Since the traditional approach using epPCR had not provided mutants displaying notable degrees of improved enantioselectivity, a different strategy was sought. At about the same time, the concept of CASTing for broadening the substrate scope of enzymes was being developed [25]. It thus appeared logical to extend it to iterative CASTing, which is an embodiment of ISM. The CAST analysis of ANEH on the basis of its X-ray structure suggested six sites for saturation mutagenesis, namely, A, B, C, D, E, and F (Figure 2.14). These sites are deﬁned by the following amino acid positions: 193/195/296(A), 215/217/219(B), 329/330(C), 349/ 350(D), 317/318(E), and 244/245/249(F). Thus, the general ISM scheme in Figure 2.3 has been extended by two sites to a total of six. The choice of the particular upward pathway in the kinetic resolution of rac-19, that is, the speciﬁc order of choosing the sites in ISM, appeared arbitrary. Indeed, the pathway B C D F E, without utilizing A, was the ﬁrst one that was chosen, and it led to a spectacular increase in enantioselectivity (Figure 2.15). The ﬁnal mutant, characterized by nine mutations, displays a selectivity factor of E ¼ 115 in the model reaction [23]. This result is all the more remarkable in that only 20 000 clones were screened, which means that no attempt was made to fully cover the deﬁned protein sequence space. Indeed, relatively small libraries were screened. The results indicate the efﬁciency of iterative CASTing and its superiority over other strategies such as repeating cycles of epPCR. It will be of interest to see whether other pathways (order of choosing the sites in ISM) also provide highly enantioselective mutants, and whether they are different. Moreover, the source of enantioselectivity needs to be illuminated in structural, kinetic, and theoretical studies. Along synthetic lines, it was observed early on that trans-disubstituted epoxides are not accepted by the WT ANEH. However, preliminary results of applying restricted CASTing on the basis of NDT instead of NNK degeneracy are highly promising N: adenine/cytosine/guanine/thymine; K: guanine/thymine; D: adenine/guanine/thymine; T: thymine [62]. The idea behind the use of NDT degeneracy has to do with the fact that only 12 amino acids are employed as building blocks, and that the screening effort for 95% coverage of the focused protein sequence space is greatly reduced (Section 2.2). Using this approach, highly active and enantioselectivity mutants were evolved [62]. This is an important synthetic result because the chiral salen-based Jacobsen catalysts fail in the case of trans-disubstituted epoxides [63]. In another study that appeared prior to the advent of CASTing, the traditional combination of epPCR and DNA shufﬂing was used to enhance the enantioselectivity of the hydrolytic kinetic resolution of p-nitrophenyl glycidyl ether and other epoxides catalyzed by the EH from Agrobacterium radiobacter [59]. Several mutants were obtained with up to 13-fold improved enantioselectivity. The amino acid exchanges took place around the active site.

2.4 Examples of Enhancing the Enantioselectivity of Enzymes by Directed Evolution

Figure 2.14 CASTing of the epoxide hydrolase from A. niger (ANEH) based on the X-ray structure of the WT [61]. (a) Defined randomization sites A–E; (b) top view of tunnel-like binding pocket showing sites A–E (blue) and the catalytically active D192 (red) [23].

j43

j 2 Directed Evolution as a Means to Engineer Enantioselective Enzymes

44

120

E = 115 (9 Mutations)

110

100

90

Selectivity factor (E)

80

E (Thr317Trp/Thr318Val) 70

60

50

40

E = 35 30

(Leu249Tyr; 244/245 stay) F

E = 24 E = 21

20

D (Cys350Val; 349 stays)

E = 14 C (Met329Pro/Leu330Tyr) B (Leu215Phe/Ala217Asp/Arg219Ser)

10

WT-ANEH Figure 2.15 Iterative CASTing in the evolution of enantioselective epoxide hydrolases as catalysts in the hydrolytic kinetic resolution of rac-19 [23].

2.4 Examples of Enhancing the Enantioselectivity of Enzymes by Directed Evolution

2.4.7 Phosphotriesterases

Many chiral phosphates and phosphonic acid esters are of value as agricultural insecticides, while others pose a threat as potential chemical warfare agents [64]. In most cases, the biological effects arise exclusively or mainly from one enantiomeric form, with the mirror image compound showing no biological activity or a reduced one [64,65]. One synthetic approach is to perform hydrolytic kinetic resolution using phosphotriesterases as catalysts, a typical example being the production of chiral organophosphates (R)- or (S)-21 (see Scheme 2.5) [65,66]. O R1O P OR2 O

H2O Phosphotriesterase

NO2 rac-21

O R1O P OR2 O

+

O R1O P OR2 OH

NO2 (R)- or (S)-21

22

Scheme 2.5

The phosphotriesterase from Pseudomonas diminuta was shown to catalyze the enantioselective hydrolysis of several racemic phosphates (21), the SP isomer reacting faster than the RP compound [65,66]. Further improvements using directed evolution were achieved by ﬁrst carrying out a restricted alanine-scan [67] (i.e. at predetermined amino acid positions alanine was introduced). Whenever an effect on activity/ enantioselectivity was observed, the position was deﬁned as a hot spot. Subsequently, randomization at several hot spots was performed, which led to the identiﬁcation of several highly (S)- or (R)-selective mutants [66]. A similar procedure was applied to the generation of mutant phosphotriesterases as catalysts in the kinetic resolution of racemic phosphonates [68]. 2.4.8 Aminotransferases

In nature, aminotransferases participate in a number of metabolic pathways [4]. They catalyze the transfer of an amino group originating from an amino acid donor to a 2-ketoacid acceptor by a simple mechanism. First, an amino group from the donor is transferred to the cofactor pyridoxal phosphate with formation of a 2-keto acid and an enzyme-bound pyridoxamine phosphate intermediate. Second, this intermediate transfers the amino group to the 2-keto acid acceptor. The reaction is reversible, shows ping-pong kinetics, and has been used industrially in the production of amino acids [69]. It can be driven in one direction by the appropriate choice of conditions (e.g. substrate concentration). Some of the aminotransferases accept simple amines instead of amino acids as amine donors, and highly enantioselective cases have been reported [70].

j45

j 2 Directed Evolution as a Means to Engineer Enantioselective Enzymes

46

O

NH2 23

NH2

(S)-transaminase O (S)-24

Scheme 2.6

However, this is obviously not always the case, a counter example being the transformation of b-tetralone (23) to (S)–aminotetralin (24), with 2-aminobutane serving as the amine donor. The WT of an (S)-selective aminotransaminase leads to a value of only 65% ee. In an important early study, a single round of epPCR was performed, and about 10 000cloneswerescreened(nodetails given)[70].Thisledtoseveralactivemutantswith improved enantioselectivity in the range of 84 –94% ee. Some had only one amino acid exchange, while others had two (e.g. M245V/R405G leading to 94% ee). This combinatorial approach does not constitute an evolutionary process, but one can expect that a second round of epPCR, DNA shufﬂing, or CASTing should result in higher enantioselectivities. In a follow-up study using a single round of epPCR, (S)-1-methoxypropan-2amine was prepared with >99% ee and 97% conversion – a nice case in which evolutionary optimization is not necessary [71]. A more recent study focused on the directed evolution of the o-transaminase from Vibrio ﬂuvialis JS17, speciﬁcally with the aim to eliminate product inhibition by aliphatic ketones while maintaining high enantioselectivity. This was achieved by screening 85 000 clones produced by epPCR [72]. 2.4.9 Aldolases

Although a host of asymmetric transition metal-catalyzed [2] and organocatalyzed [3] aldol reactions are known, many problems persist. Aldolases are often complementary, allowing the enantio- and diastereoselective synthesis of many complex carbohydrates in one or two steps [4,73]. Of course, as in other WTenzyme-catalyzed processes, broad substrate acceptance and stereoselectivity cannot be expected to be general. Thus, directed evolution offers intriguing prospects. The ﬁrst such study addressed the limited substrate acceptance of D-2-keto-3-deoxy-6-phospho-gluconate (KDPG) aldolase [74]. This enzyme catalyzes the (reversible) addition of pyruvate (25) to certain chiral aldehydes such as (26), with formation of aldol products such as (27). It was known that this aldolase is highly speciﬁc for chiral phosphorylated aldehydes with the D conﬁguration at the C2 position leading stereoselectively to a precursor of the corresponding D sugar such as (28) (see Scheme 2.7) [75]: Unfortunately, the phosphorylated form of the starting aldehyde is expensive, and dephosphorylation by a phosphatase requires an additional step. Therefore, the challenge was to obtain a mutant aldolase that not only accepts nonphosphorylated substrates but also turns over the enantiomeric aldehyde (29) stereoselectively with formation of (30), which is a precursor of carbohydrate (31) (see Scheme 2.8) [74]:

2.4 Examples of Enhancing the Enantioselectivity of Enzymes by Directed Evolution

-O

2C

+

OHC

O

KDPG aldolase

OPO32-

-O

OH OPO32-

2C

OH

O 25

j47

OH

26

Phosphatase

27 O

HO OH

CO2-

OH

28 Scheme 2.7

It was indeed possible to turn the KDPG aldolase into an efﬁcient mutant that accepts aldehyde (29), while maintaining perfect control over the creation of the new stereocenter. A few rounds of epPCR and DNA shufﬂing led to several highly active and stereoselective mutants. One of the best mutants turned out to have four mutations (T84A, I92F, V118A, E138V), most of which occur at positions remote from the active site [74]. A detailed explanation has not been proposed, but this investigation once more shows that directed evolution may lead to effects not predicted by current rational design. It would be interesting to see how iterative CASTing focusing on the binding pocket would perform. This work was extended by subjecting the Neu5Ac aldolase from E. coli to directed evolution in order to expand its catalytic activity for enantiomeric forms of the usual substrates, including N-acetyl-L-mannosamine and L-arabinose with formation of the synthetically important products L-sialic acid and L-3-deoxymanno-2-octulosonic acid (L-KDO) [76]. The evolved Neu5Ac aldolases were characterized by sequence analysis, kinetics, stereoselectivity, and in one case even by an X-ray structure analysis. Again, remote mutations were identiﬁed. Surprisingly, the X-ray structure of one of the triple mutants at 2.3 A resolution showed no signiﬁcant difference in folding, relative to that of the wild type. The Wong group has extended their studies to include the directed evolution of D-sialic acid aldolase to L-3-deoxy-manno-2-octulosonic acid (L-KDO) aldolase, which likewise underscores the signiﬁcance of this approach [77a]. In a similar approach, the industrial synthesis of (3R,5S)-6-chloro-2,4,6-trideoxyhexapyranoside has been accomplished by using a genetically altered aldolase [77b]. In another intriguing directed evolution study, the stereochemical course of aldol additions was signiﬁcantly altered in a different sense [78]; rather than evolving aldolase mutants that selectively accept stereoisomers of substrates, the —O

2C

O

OHC

OH

+ O 25

Scheme 2.8

OH 29

—O

OH

OH OH

2C

OH 30

O

HO HO

31

CO2-

j 2 Directed Evolution as a Means to Engineer Enantioselective Enzymes

48

OH Wild type O

2—O

OPO32—

3PO

OH

OPO32—

HO

O

OH

33

32 Directed evolution OHC

OPO32—

OH

OH 26

Mutant

2—O

O OPO32—

3PO

OH

OH

34 Scheme 2.9

conﬁguration of the chiral aldehyde as substrate was kept constant and mutants were evolved, which showed the opposite diastereoselectivity (see Scheme 2.9). This goal was reached for the ﬁrst time in a directed evolution study involving the so-called tagatose-1,6-bisphosphate aldolase [78]. The WT catalyzes the aldol addition of (32) to (26) with selective formation of aldol adduct (33). Upon applying three rounds of DNA shufﬂing and screening 8000 mutants, an aldolase was obtained showing an 80-fold improvement of kcat/Km of the reaction 26 þ 32 ! 34. Thus, a tagatose aldolase was turned into a fructose aldolase. Investigations of this kind are of great value to synthetic organic chemists interested in the selective synthesis of complex stereoisomeric products such as carbohydrates. 2.4.10 Cyclohexanone and Cyclopentanone Monooxygenases as Baeyer–Villigerases and Sulfoxidation Catalysts

Chemo-, regio-, and stereoselective partial oxidation is an area of intense interest in synthetic organic chemistry [1,2,79,80]. Synthetic transition metal catalysts and enzymes have already contributed successfully, but major challenges remain. The Baeyer–Villiger (BV) reaction of ketones with hydroperoxides affording the corresponding esters or lactones constitutes a synthetically important partial oxidation [81–83]. The reaction is accelerated by acids, bases, or metal complexes, with enantioselective catalysis also being possible in some cases. Alternatively, ﬂavindependent enzymes of the type cyclohexanone monooxygenases (CHMO) can be used as biocatalysts in asymmetric BV reactions [84–87]. In this enzymatic process, oxygen (in air) reacts with the enzyme-bound ﬂavin (FAD) to form an intermediate hydroperoxide, which in the deprotonated form initiates the BV reaction by transferring one oxygen atom originating from O2 to the substrate (ketone), the other being ultimately reduced to water [88]. The ﬂavin cofactor needs to be recycled by reduction, the second cofactor NADPH taking over this function (Figure 2.16).

R

N

N N

NADPH

H

H

O

O2

H2 O

N

H+

N O H - O O

N

R

N H

O R1

O Figure 2.16 Mechanism of CHMO-catalyzed Baeyer–Villiger reaction [85,88].

N

N

O

N H

O

NADP+

N-

N

R

O

N N OH O

N

R

R2

N

N

R2

N H

O

H

O

R1

O

OCriegee-intermediate

R1

N O H O O

N

R

O R2

2.4 Examples of Enhancing the Enantioselectivity of Enzymes by Directed Evolution

j49

j 2 Directed Evolution as a Means to Engineer Enantioselective Enzymes

50

Although NADPH recycling systems are well known, it is currently more convenient to use whole cells containing CHMO. CHMO is known to catalyze a number of enantioselective BV reactions, including the kinetic resolution of certain racemic ketones and desymmetrization of prochiral substrates [84–87]. An example is the desymmetrization of 4-methylcyclohexanone, which affords the (S)-conﬁgurated seven-membered lactone with 98% ee [84,87]. Of course, many ketones fail to react with acceptable levels of enantioselectivity, or are not even accepted by the enzyme. The initial results of an early directed evolution study are all the more signiﬁcant, because no X-ray data or homology models were available then to serve as a possible guide [89]. In a model study using whole E. coli cells containing the CHMO from Acinetobacter sp. NCIMB9871, 4-hydroxy-cyclohexanone(35) was usedas thesubstrate. The WT leads to the preferential formation of the primary product (R)-36, which spontaneously rearranges to the thermodynamically more stable lactone (R)-37. The enantiomeric excess of this desymmetrization is only 9%, and the sense of enantioselectivity (R) is opposite to the usually observed (S)-preference displayed by simple 4-alkyl-substituted cyclohexanone derivatives (see Scheme 2.10) [84–87]. In the directed evolution study, epPCR at various mutation rates was applied (10 000 clones). Some of the hits were (R) selective, and others were (S) selective. Eight of them were sequenced [89]. Of particular interest is mutant 1-K2-F5, characterized by a single mutation F432S, because it leads to reversal of enantioselectivity (79% ee in favor of (S)-37). (R) selectivity was optimized by using the genes of some of the best (R)-selective mutants as starting points for a second round of epPCR [89]. In each case only about 1600 mutants were screened. Although a systematic search was not a goal of this initial investigation, the strategy turned out to be successful. In the case of mutant 1-F1-F5 (40% ee), characterized by a single mutation L143F, a second round of epPCR led to a markedly improved mutant 2-D19-E6, showing 90% ee in favor of (R)-37. Sequencing revealed that four new amino acid exchange events had occurred (E292G, T433I, L435Q, T464A) in addition to the already existing L143F mutation.

O

O

H O

O O2

OH

WT CHMO or mutants

OH

O

35

O OH (S)-36

Scheme 2.10

OH

(R)-37

HO (R)-36 O

O

O (S)-37

H

2.4 Examples of Enhancing the Enantioselectivity of Enzymes by Directed Evolution

O O2

O

38

O

O

+

CHMOmutant OCH3

O

OCH3 (R)-39

OCH3 (S)-39

Scheme 2.11

As the WT CHMO was known to react (S) selectively with simple four-substituted cyclohexanone derivatives [84–87], it was logical to test mutant 1-K2-F5 as a catalyst in the BV reaction of other ketones. For example, when 4-methoxycyclohexanone (38) was subjected to the BV reaction catalyzed by mutant 1-K2-F5, almost complete enantioselectivity was observed in favor of the (S)-lactone (39) (98.5% ee), in contrast to the WT, which is considerably less selective (78% ee) (see Scheme 2.11) [89]. When using mutant 1-K2-F5, other four-substituted cyclohexanone derivatives such as the methyl, ethyl, chloro, bromo, and iodo compounds were found to react with enantioselectivities of 95–99% ee [89]. These enantiomeric excess values are similar to those of the WT, which shows that the mutational change F432S does not simply shift, but rather expands the substrate range in which >95% ee is achieved. Upon testing a fairly wide range of other substrates including cyclopentanone derivatives, bicyclic ketones, functionalized cyclohexanone, and cyclobutanone derivatives, it was discovered that mutant 1-K2-F5 is astonishingly effective [90] (Table 2.1). Enantioselectivities of 90–99% were generally observed. No synthetic catalyst known to date can compete [82,83]. Shortly after the publication of the early work [89], a paper appeared describing the ﬁrst X-ray structure of a Baeyer–Villigerase, namely, phenylacetone monooxygenase (PAMO) [91]. It was originally found in the thermophilic bacterium Thermobiﬁda fusca and is the ﬁrst thermostable Baeyer–Villigerase [92]. Unfortunately, it is characterized by narrow range of substrate acceptance. The X-ray data of PAMO allowed the construction of a homology model for CHMO and the possibility of interpreting the previous mutational results on a molecular level [93]. Although the theoretical analysis has not been completed, a cursory inspection suggests that the newly introduced serine at position 432 of mutant 1-K2-F5 forms a hydrogen bond with arginine at position 337, the amino acid which was postulated to stabilize the Criegee intermediate. Since PAMO is unusually thermostable [91,92], it seemed attractive to engineer a mutant showing broader substrate acceptance [93]. Thus, PAMO was turned into a phenylcyclohexanone monooxygenase (PCHMO) by rational design [93]. The best mutant accepts 2-phenylcyclohexanone, inter alia, and leads to an E value of 100 in kinetic resolution, while retaining high thermal stability. The WT CHMO also shows high enantioselectivity in this reaction, but it is not very stable thermally. This investigation constitutes an important step toward developing robust (and thus industrially viable) Baeyer–Villigerases. Moreover, several parameters in a large scale in vitro reaction were optimized, including buffer composition, solvent and cofactor

j51

j 2 Directed Evolution as a Means to Engineer Enantioselective Enzymes

52

Table 2.1 Desymmetrization of prochiral ketones by the BV

reaction using O2 as the oxidant and the CHMO mutant 1-K2-F5 as the catalyst [90]. Desymmetrization

ee (%) 94

O O

O O Cl

O

Cl

O 91

O O

99

O

O O

97

O 78

O

O O O

O

O

O

O

O

O

O

OH

96

>99

O

>99

O

O

>99

O

OH

regeneration system [94]. A light-driven BV reaction showing >95% ee was also devised [95]. Another FAD/NADPH-dependent enzyme is cyclopentanone monooxygenase (CPMO), which appears to have a somewhat different catalytic proﬁle [96]. For this reason a directed evolution project was initiated based on CASTing, although only very small libraries were screened in the early phase of this research (mini directed evolution) [26]. Codon usage was restricted to NDTdegeneracy, thus utilizing only 12 amino acids. The added challenge in applying CASTing in this case has to do with the dynamics of this class of enzymes. Large conformational changes occur upon

2.4 Examples of Enhancing the Enantioselectivity of Enzymes by Directed Evolution

H3C

H3C O2 CH3

H2C

40

S

O

..

H2 C

S

+

O

CHMOmutants

CH3

(R)-41

H2C

S

..

H3 C

CH3

(S)-41

Scheme 2.12

substrate binding. Nevertheless, on the basis of the X-ray structure of PAMO and sequence alignments, two sites were identiﬁed in CPMO for CAST libraries, namely, F156/G157 (library A) and G449/F450 (library B). Saturation mutagenesis at these CAST sites using NDT degeneracy and screening only 150 clones (65% coverage) in each case led to several mutants, which show high enantioselectivity in several BV reactions. For example, in the case of 4-methylcyclohexanone, the WT CPMO shows a value of only 46% ee (R), while a mutant boosted enantioselectivity to 92% ee (R). Iterative CASTing [23] can be expected to deliver even better results. It had been known for a long time that CHMO can also be used as catalysts in the enantioselective air-oxidation of some but not all prochiral thioethers [97]. Therefore, directed evolution of the CHMO from Acinetobacter sp. NCIMB 9871 was applied in those cases in which the WT fails [32]. An example is substrate (40), which reacts with an enantiomeric excess of only 14% in slight favor of (R)-41 (see Scheme 2.12). The original library of 10 000 clones used in the Baeyer–Villiger reaction [89] was screened for performance as potential catalysts in the sulfoxidation [32]. This led to the discovery of at least 20 mutants having enantiomeric excess values in the range of 85–99%, some being (R) selective and others being (S) selective. Five mutants resulting in enantiomeric excess values of >95% were sequenced (Table 2.2) [32]. In general, the mutants are different from those that were previously identiﬁed as hits in the BV reaction of prochiral cyclohexanone derivatives, which is not surprising. In contrast, mutant 1-K15-C1, which leads to 98.7% ee in favor of (R)-41, is characterized by amino acid exchange F432S. It is therefore identical to mutant Table 2.2 Selected mutant CHMOs from Acinetobacter sp. NCIMB 9871 for the enantioselective air-oxidation of thio-ether 40 (24 hours; 23–25 C) using whole cells [32].

Mutant

Amino acid exchanges

WT 1-D10-F6 1-K15-C1 1-C5-H3 1-H8-A1

— D384H F432S K229I, L248P Y132C, F246I, V361A, T415A F16L, F277S

1-J8-C5

Yield of 41 (%)

Configuration

ee (%)

Sulfone as side product (%)

75 75 55 77 52

(R) (R) (R) (S) (S)

14.0 98.9 98.7 98.1 99.7

<1 7.9 20.0 5.6 26.6

84

(S)

95.2

5.6

j53

j 2 Directed Evolution as a Means to Engineer Enantioselective Enzymes

54

1-K2-F5 previously identiﬁed as a highly enantioselective biocatalyst in the asymmetric BV reaction of a wide range of four-substituted cyclohexanone derivatives (as shown above). Thus, one and the same single mutational change at position 432, namely, the exchange of phenylalanine by serine, leads to a surprisingly versatile biocatalyst [32]. Although a detailed mechanistic study still needs to be performed, S432 may well form a H-bond to the crucial amino acid R337. In the sulfoxidation, appreciable over-oxidation with formation of undesired sulfone was observed [32]. Directed evolution was then applied successfully to eliminate undesired sulfone formation, speciﬁcally by going through a second cycle of epPCR [32]. This is signiﬁcant because it shows for the ﬁrst time that an undesired side reaction can be eliminated by directed evolution. 2.4.11 Monoamine Oxidases

Monoamine oxidases are enzymes that catalyze the racemization of -amino acids [98]. Both L- and D-selective monoamine oxidases are known (i.e. they catalyze the racemization of either (S) or (R) enantiomers of amino acids). This property has been exploited to obtain enantiomerically pure (R) and (S) amino acids by using an appropriate achiral reducing agent such as NaBH4, NaB(CN)H3 or H3NBH3 in combination with an L- or D-selective monoamine oxidase [99]. This synthetically interesting procedure was extended to the production of chiral amines such as a-methylbenzylamine (42) [100]. Unfortunately, the enzymes display somewhat low activity toward substrates of this kind. For example, the monoamine oxidase from A.niger (MAO-N) racemizes (42), with kcat being only 0.17 min1, although enantioselectivity is not bad ((S) versus (R) selectivity amounts to 17: 1) [100]. Thus, directed evolution was applied (see Scheme 2.13). Following several cycles of mutagenesis using the E. coli XL1-Red mutator strain and transformation of the plasmid library into E. coli, a total of about 150 000 bacterial colonies were assayed for activity using a colorimetric prescreen [100]. The best mutant Asn336Ser showed a 47-fold increase in activity and a 5.8-fold enhancement

NH2 Ph

CH3

(S)-selective MAO

(S)-42 NH H3N . BH3 NH2 Ph

CH3

(R)-42 Scheme 2.13

Ph

CH3 43

2.4 Examples of Enhancing the Enantioselectivity of Enzymes by Directed Evolution

of (S) enantioselectivity [100]. Later it was shown that this mutant displays a fairly broad range of primary and secondary amines [101]. More recently, the chemoenzymatic approach has been extended to include the difﬁcult class of tertiary amines [102]. These results illustrate the synthetic power of this approach, which is currently being used industrially. 2.4.12 Cytochrome P450 Enzymes

Cytochrome P450 enzymes occur widely in nature, catalyzing the O2-driven oxidation of an immense variety of organic substrates [4,103]. The mechanism of these Fe-heme-containing enzymes has been investigated extensively, and many have been used as whole-cell catalysts in organic chemistry, as in the CH-activating hydroxylation of steroids [104]. Directed evolution has been applied to the bacterial cytochrome P450 BM-3 from Bacillus megaterium with the aim of turning it into practical mutants showing high activity and regioselectivity toward simple substrates including linear alkanes [105]. For example, using n-octane, previous work had shown that a mixture of 2-, 3-, and 4-octanols is obtained, whereas a mutant was shown to lead to approximately 80% regioselectivity in favor of 2-octanol [105c]. The enantiopurity turned out to be only 40% ee (S), but enantioselectivity was not the goal of the investigation. In a later study, some of the previous mutants were screened as catalysts for the hydroxylation of oxazolines, which led to enantioselectivities in the range of 84–88% ee [106]. A few of the original mutants were also tested in the hydroxylation of aryl-acetic acid esters (44), which led to enantiomeric excesses of 56–95% (see Scheme 2.14) [107]. In general, it would be interesting to exploit iterative CASTing or some other form of directed evolution in the quest to control enantioselectivity of P450 enzymes. Thus far, P450 BM-3 was subjected to directed evolution in the successful quest to improve the enantioselectivity of the epoxidation of 1-hexene (up to 83% ee) [108]. Moreover, a triple mutant of P450 BM-3, originally evolved to turn it into an indolehydroxylating catalyst, was found to catalyze the regioselective hydroxylation of highly branched fatty acids [108]. 2.4.13 Other Enzymes

Several other enzymes have been investigated by directed evolution in order to control substrate speciﬁcity, although most of the studies are at an early stage. These

ArCH2CO2R

44 Scheme 2.14

P450 BM-3

Mutant

OH ArCHCO2R 45

j55

j 2 Directed Evolution as a Means to Engineer Enantioselective Enzymes

56

include the engineering of the important class of glycosidases/glycosyl transferases [109] and reductases [110].

2.5 Conclusions and Perspectives

Directed evolution of enantioselective enzymes has emerged as a fundamentally new approach to asymmetric catalysis in synthetic organic chemistry, which involves the proper combination of gene mutagenesis, expression, and high-throughput analysis of enantiomeric purity. Several industrial examples of directed evolution of enantioselective and robust enzymes have been described, others are not directly accessible in the open literature. A range of mutagenesis methods are available, and strategies have been developed for applying them when exploring protein sequences. Most directed evolution projects begin with epPCR or DNA shufﬂing, but the systematic generation of focused libraries has attracted attention recently. Accordingly, iterative CASTing, which is an example of ISM, offers interesting prospects with respect to the enlargement of substrate acceptance and the enhancement of enantioselectivity. ISM has also been applied in the quest to enhance the thermostability of enzymes substantially. Future work will focus on further generalization to include enzymes that are not yet investigated. Enzyme promiscuity is also an emerging ﬁeld [111], and directed evolution thereof is another challenge, especially when it includes enantioselectivity. Theoretical and computational approaches, including genetic algorithms or neural networks, may prove to be helpful in speciﬁc aspects of directed evolution [112]. Finally, the idea of directed evolution of enantioselective hybrid catalysts comprising a synthetic transition metal center anchored covalently or noncovalently to a protein host also offers intriguing prospects, because this allows the tuning of synthetic catalysts by the methods of molecular biology in a Darwinian sense [8b,8c]. Proof-ofprinciple using iterative CASTing was recently presented for the ﬁrst time, speciﬁcally in the case of Rh-catalyzed oleﬁn-hydrogenation [27]. The enantiomeric excess was shown to increase stepwise from 23–65% ee. This work provides incentive for designing and implementing improved systems.

References 1 (a) Collins, A.N., Sheldrake, G.N. and Crosby, J. (eds)(1997) Chirality in Industry II: Developments in the Commercial Manufacture and Applications of Optically Active Compounds, John Wiley & Sons, Ltd, Chichester. (b) Rouhi, A.M. (2003) Chemical and Engineering News, 81, 45–55. 2 Jacobsen, E.N., Pfaltz, A. and Yamamoto, H. (eds)(1999) Comprehensive Asymmetric Catalysis, Springer, Berlin, Vol. I–III.

3 Berkessel, A. and Gr€oger, H. (eds)(2004) Asymmetric Organocatalysis, Wiley-VCH Verlag GmbH, Weinheim. 4 (a) Drauz, K. and Waldmann, H. (eds) (2002) Enzyme Catalysis in Organic Synthesis: A Comprehensive Handbook, 2nd edn., Wiley-VCH Verlag GmbH, Weinheim. Vol. I–III. (b) Faber, K. (eds) (2000) Biotransformations in Organic Chemistry, 4th edn., Springer, Berlin.

References (c) Liese, A., Seelbach, K. and Wandrey, C. (eds)(2006) Industrial Biotransformations, Wiley-VCH Verlag GmbH, Weinheim. (d) Schmid, A., Dordick, J.S., Hauer, B., Kiener, A., Wubbolts, M. and Witholt, B. (2001) Nature (London U.K.), 409, 258–268. (e) Koeller, K.M. and Wong, C.-H. (2001) Nature (London U.K.), 409, 232–240. 5 Schultz, P.G., Yin, J. and Lerner, R.A. (2002) Angewandte Chemie, 114, 4607–4618. (2002) Angewandte Chemie International Edition, 41, 4427–4437. 6 (a) Reetz, M.T., Zonta, A., Schimossek, K., Liebeton, K. and Jaeger, K.-E. (1997) Angewandte Chemie, 109, 2961–2963. (1997) Angewandte Chemie International Edition in English, 36, 2830–2832. (b) Reetz M.T., Zonta A., Schimossek K., Liebeton K. and Jaeger K.-E. (1997) patent application DE-A 19731990.4 (A Process for the Preparation and Identiﬁcation of Novel Hydrolases Having Improved Properties). 7 (a) Arnold, F.H. and Georgiou, G. (eds) (2003) Directed Enzyme Evolution: Screening and Selection Methods, Humana Press, Totowa. Vol. 230. (b) Brakmann, S. and Johnsson, K. (eds)(2002) Directed Molecular Evolution of Proteins (or How to Improve Enzymes for Biocatalysis), WileyVCH Verlag GmbH, Weinheim. (c) Brakmann, S. and Schwienhorst, A. (eds) (2004) Evolutionary Methods in Biotechnology, (Clever Tricks for Directed Evolution), Wiley-VCH Verlag GmbH, Weinheim. (d) Taylor, S.V., Kast, P. and Hilvert, D. (2001) Angewandte Chemie, 113, 3408–3436. (2001) Angewandte Chemie International Edition, 40, 3310–3335. (e) Powell, K.A., Ramer, S.W., del Cardayre, S.B., Stemmer, W.P.C., Tobin, M.B., Longchamp, P.F. and Huisman, G.W. (2001) Angewandte Chemie, 113, 4068–4080; (2001) Angewandte Chemie International Edition, 40, 3948–3959. (f) Rubin-Pitel, S.B. and Zhao, H. (2006) Combinatorial Chemistry and High Throughput Screening, 9, 247–257. (g) Kaur, J. and Sharma, R.

8

9

10

11

(2006) Critical Reviews in Biotechnology, 26, 165–199. (a) Reetz, M.T. (1999) Pure and Applied Chemistry, 71, 1503–1509. (b) Reetz, M.T. (2002) Tetrahedron, 58, 6595–6602. (c) Reetz, M.T. (2004) Proceedings of the National Academy of Sciences of the United States of America, 101, 5716–5722. (d) Reetz, M.T. (2004) Methods in Enzymology, (eds D.E. Robertson and J.P. Noel), Elsevier Academic Press, San Diego, Vol. 388, pp. 238–256. (a) Imanaka, T. and Atomi, H. (2002) In Enzyme Catalysis in Organic Synthesis: A Comprehensive Handbook, 2nd edn (eds K. Drauz and H. Waldmann), Wiley-VCH Verlag GmbH, Weinheim, Vol. 1, pp. 67–95. (b) Cedrone, F., Menez, A. and Quemeneur, E. (2000) Current Opinion in Structural Biology, 10, 405–410. (c) Harris, J.L. and Craik, C.S. (1998) Current Opinion in Chemical Biology, 2, 127–132. (d) Kazlauskas, R.J. (2000) Current Opinion in Chemical Biology, 4, 81–88. (e) Li, Q.-S., Schwaneberg, U., Fischer, M., Schmitt, J., Pleiss, J., Lutz-Wahl, S. and Schmid, R.D. (2001) Biochimica et Biophysica Acta, 1545, 114–121. See for example: (a) Cunningham, B.C. and Wells, J.A. (1987) Protein Engineering, 1, 319–325. (b) Lehtovaara, P.M., Koivula, A.K., Bamford, J. and Knowles, J.K.C. (1988) Protein Engineering, 2, 63–68. (c) Liao, H., McKenzie, T. and Hageman, R. (1986) Proceedings of the National Academy of Sciences of the United States of America, 83, 576–580. (a) Chen, K. and Arnold, F.H. (1993) Proceedings of the National Academy of Sciences of the United States of America, 90, 5618–5622. (b) Arnold, F.H. (1998) Accounts of Chemical Research, 31, 125–131. (c) Stemmer, W.P.C. (1994) Nature (London), 370, 389–391. (d) Stemmer, W.P.C. (1994) Proceedings of the National Academy of Sciences of the United States of America, 91, 10747–10751. (e) Reetz, M.T. and Jaeger, K.-E. (1999) Topics in Current Chemistry, 200, 31–57.

j57

j 2 Directed Evolution as a Means to Engineer Enantioselective Enzymes

58

12 Lutz, S. and Patrick, W.M. (2004) Current Opinion in Biotechnology, 15, 291–297. 13 (a) For an illuminating comparison of directed evolution approaches using the glucuronidase model system, see: Rowe, L.A., Geddie, M.L., Alexander, O.B. and Matsumura, I. (2003) Journal of Molecular Biology, 332, 851–860. (b) Study of saturation mutagenesis versus DNA shufﬂing:Parikh, M.R. and Matsumura, I. (2005) Journal of Molecular Biology, 352, 621–628. (c) Statistical analysis of random mutagenesis methods: Wong, T.S., Roccatano, D., Zacharias, M. and Schwaneberg, U. (2006) Journal of Molecular Biology, 355, 858–871. 14 Leung, D.W., Chen, E. and Goeddel, D.V. (1989) Technique (Philadelphia), 1, 11–15. 15 Cadwell, R.C. and Joyce, G.F. (1992) PCRMethods and Applications, 2, 28–33. 16 (a) Hughes, M.D., Nagel, D.A., Santos, A.F., Sutherland, A.J. and Hine, A.V. (2003) Journal of Molecular Biology, 331, 973–979. (b) Cirino, P.C., Mayer, K.M. and Umeno, D. (2003) Directed Evolution Library Creation: Methods and Protocols (eds F.H. Arnold and G. Georgiou), Humana Press, Totowa, Vol. 231, pp. 3–10. (c) Eggert, T., Reetz, M.T. and Jaeger, K.-E. (2004) Enzyme Functionality - Design, Engineering and Screening (ed. A. Svendsen), Marcel Dekker, New York, pp. 375–390. (d) Wong, T.S., Tee, K.L., Hauer, B. and Schwaneberg, U. (2004) Nucleic Acids Research, 32, e26. 17 (a) Georgescu, R., Bandara, G. and Sun, L. (2003) Directed Evolution Library Creation (eds F.H. Arnold and G. Georgiou), Humana Press, Totowa, pp. 75–84. (b) Wells, J.A., Vasser, M. and Powers, D.B. (1985) Gene, 34, 315–323. (c) Barettino, D., Feigenbutz, M., Valcarcel, R. and Stunnenberg, H.G. (1994) Nucleic Acids Research, 22, 541–542. 18 (a) Bosley, A.D. and Ostermeier, M. (2005) Biomolecular Engineering, 22, 57–61. (b) Patrick, W.M. and Firth, A.E. (2005) Biomolecular Engineering, 22, 105–112. 19 (a) Funke, S.A., Eipper, A., Reetz, M.T., Otte, N., Thiel, W., van Pouderoyen, G.,

20

21

22

23

24 25

26

27

28

Dijkstra, B.W., Jaeger, K.-E. and Eggert, T. (2003) Biocatalysis and Biotransformation, 21, 67–73.(b) Krumm, H. (2004) Dissertation, Ruhr-Universit€at Bochum, Germany. DeSantis, G., Wong, K., Farwell, B., Chatman, K., Zhu, Z., Tomlinson, G., Huang, H., Tan, X., Bibbs, L., Chen, P., Kretz, K. and Burk, M.J. (2003) Journal of the American Chemical Society, 125, 11476–11477. (a) Miyazaki, K. and Arnold, F.H. (1999) Journal of Molecular Evolution, 49, 716–720. (b) Sneeden, J.L. and Loeb, L.A. (2003) Directed Evolution Library Creation: Methods and Protocols (eds F.H. Arnold and G. Georgiou), Humana Press, Totowa, Vol. 231 pp. 65. literature cited therein; (c) Liebeton, K., Zonta, A., Schimossek, K., Nardini, M., Lang, D., Dijkstra, B.W., Reetz, M.T. and Jaeger, K.-E. (2000) Chemistry and Biology, 7, 709–718. Reetz, M.T., Wilensek, S., Zha, D. and Jaeger, K.-E. (2001) Angewandte Chemie, 113, 3701–3703; (2001) Angewandte Chemie International Edition, 40, 3589–3591. Reetz, M.T. and Wang, L.-W. (2006) Bocola in part M. Angewandte Chemie, 118, 1258–1263. Erratum, 2556 (2006) Angewandte Chemie International Edition, 45, 1236–1241. Erratum, 2494. Reetz, M.T. and Carballeira, J.D. Nature Protocols, submitted. Reetz, M.T., Bocola, M., Carballeira, J.D., Zha, D. and Vogel, A. (2005) Angewandte Chemie, 117, 4264–84268 (2005) Angewandte Chemie International Edition, 44, 4192–4196. Clouthier, C.M., Kayser, M.M. and Reetz, M.T. (2006) Journal of Organic Chemistry, 71, 8431–8437. Reetz, M.T., Peyralans, J.J.-P., Maichele, A., Fu, Y. and Maywald, M. (2006) Chemical Communications (Cambridge, U.K.), 4318–4320. Reetz, M.T., Carballeira, J.D. and Vogel, A. (2006) Angewandte Chemie, 118, 7909–7915. (2006) Angewandte Chemie

References

29

30

31

32

33

34

International Edition, 45, 7745–7751. Crameri, A., Raillard, S.-A., Bermudez, E. and Stemmer, W.P.C. (1998) Nature (London U.K.), 391, 288–291. (a) Reetz, M.T. (2001) Angewandte Chemie, 113, 292–320. (2001) Angewandte Chemie International Edition, 40, 284–310.(b) Reetz, M.T. (2003) Methods in Molecular Biology (eds F.H. Arnold and G. Georgiou), Humana Press, Totowa, Vol. 230, pp. 259–282. (c) Reetz, M.T. (2003) Methods in Molecular Biology (eds F.H. Arnold and G. Georgiou), Humana Press, Totowa, Vol. 230, pp. 283–290. (d) Reetz, M.T. (2004) Evolutionary Methods in Biotechnology (eds S. Brakmann and A. Schwienhorst), Wiley-VCH Verlag GmbH, Weinheim, pp. 113–141. (e) Wahler, D. and Reymond, J.-L. (2001) Current Opinion in Biotechnology, 12, 535–544. (f) Schmidt, M. and Bornscheuer, U.T. (2005) Biomolecular Engineering, 22, 51–56. (g) Reymond, J-.L. (ed.)(2005) Enzyme Assays - Highthroughput Screening, Genetic Selection and Fingerprinting, Wiley-VCH Verlag GmbH, Weinheim. Reetz, M.T., K€ uhling, K.M., Wilensek, S., Husmann, H., H€ausig, U.W. and Hermes, M. (2001) Catalysis Today, 67, 389–396. Reetz, M.T., Daligault, F., Brunner, B., Hinrichs, H. and Deege, A. (2004) Angewandte Chemie, 116, 4170–4173 (2004) Angewandte Chemie International Edition, 43, 4078–4081. Review: Reetz, M.T. (2006) Advances in Catalysis (eds B.C. Gates and H. Kn€ ozinger), Elsevier, San Diego, Vol. 49, pp. 1–69. (a) Schmid, R.D. and Verger, R. (1998) Angewandte Chemie, 110, 1694–1720; (1998) Angewandte Chemie International Edition, 37, 1608–1633. (b) Krishna, S.H. and Karanth, N.G. (2002) Catalysis Reviews, 44, 499–591. (c) Berglund, P. and Hult, K. (2000) Stereoselective Biocatalysis (eds R.N. Patel and N. Ramesh), Marcel

35

36

37 38

39 40 41

42

43

44

Dekker, New York, Chapter 21, pp. 633–657. (d) QM-Studies of serine hydrolases: Hu, C.-H., Brinck, T. and Hult, K. (1997) International Journal of Quantum Chemistry, 89–103. (e) Mechanism of serine proteases: Hedstrom, L. (2002) Chemical Reviews, 102, 4501–4523. Jaeger, K.-E., Ransac, S., Dijkstra, B.W., Colson, C., van Heuvel, M. and Misset, O. (1994) FEMS Microbiology Review, 15, 29–63. (a) Bocola, M., Otte, N., Jaeger, K.-E., Reetz, M.T. and Thiel, W. (2004) Chembiochem: A European Journal of Chemical Biology, 5, 214–223. (b) Reetz, M.T., Puls, M., Carballeira, J.D., Vogel, A., Jaeger, K.-E., Eggert, T., Thiel, W., Bocola, M. and Otte, N. (2007) Chembiochem: A European Journal of Chemical Biology, 8, 106–112. Reetz, M.T. (2000) Pure and Applied Chemistry, 72, 1615–1622. Arnold, F.H., Wintrode, P.L., Miyazaki, K. and Gershenson, A. (2001) Trends in Biochemical Sciences, 26, 100–106. Crameri, A. and Stemmer, W.P.C. (1995) BioTechniques, 18, 194–196. Reetz, M.T. and Jaeger, K.-E. (2000) Chemistry-A European Journal, 6, 407–412. Zha, D., Wilensek, S., Hermes, M., Jaeger, K.-E. and Reetz, M.T. (2001) Chemical Communications (Cambridge, U.K.), 2664–2665. Nardini, M., Lang, D.A., Liebeton, K., Jaeger, K.-E. and Dijkstra, B.W. (2000) Journal of Biological Chemistry, 275, 31219–31225. Ifﬂand, A., Tafelmeyer, P., Saudan, C. and Johnsson, K. (2000) Biochemistry, 39, 10790–10798. (a) Kumar, M., Kannan, K.K., Hosur, M.V., Bhavesh, N.S., Chatterjee, A., Mittal, R. and Hosur, R.V. (2002) Biochemical and Biophysical Research Communications, 294, 395–401. (b) Agarwal, P.K., Billeter, S.R., Rajagopalan, P.T.R., Benkovic, S.J. and Hammes-Schiffer, S. (2002) Proceedings of the National Academy of

j59

j 2 Directed Evolution as a Means to Engineer Enantioselective Enzymes

60

45

46

47

48

49 50

51

52

53

54

55 56

Sciences of the United States of America, 99, 2794–2799. Fersht, A. (1999) Structure and Mechanism in Protein Science, W.H. Freeman, New York. Reetz, M.T., Carballeira, J.D., Peyralans, J.J.-P., H€obenreich, H., Maichele, A. and Vogel, A. (2006) Chemistry-A European Journal, 12, 6031–6038. Zha, D., Eipper, A. and Reetz, M.T. (2003) Chembiochem: A European Journal of Chemical Biology, 4, 34–39. Koga, Y., Kato, K., Nakano, H. and Yamane, T. (2003) Journal of Molecular Biology, 331, 585–592. Moore, J.C. and Arnold, F.H. (1996) Nature Biotechnology, 14, 458–467. (a) Bornscheuer, U.T., Altenbuchner, J. and Meyer, H.H. (1998) Biotechnology and Bioengineering, 58, 554–559. (b) Henke, E. and Bornscheuer, U.T. (1999) Biological Chemistry, 380, 1029–1033. (a) Horsman, G.P., Liu, A.M.F., Henke, E., Bornscheuer, U.T. and Kazlauskas, R.J. (2003) Chemistry-A European Journal, 9, 1933–1939. (b) Park, S., Morley, K.L., Horsman, G.P., Holmquist, M., Hult, K. and Kazlauskas, R.J. (2005) Chemistry and Biology, 12, 45–54. Schmidt, M., Hasenpusch, D., K€ahler, M., Kirchner, U., Wiggenhorn, K., Langel, W. and Bornscheuer, U.T. (2006) Chembiochem: A European Journal of Chemical Biology, 7, 805–809. Zhang, G., Gao, R., Zheng, L., Zhang, A., Wang, Y., Wang, Q., Feng, Y. and Cao, S. (2006) Journal of Molecular Catalysis B: Enzymatic, 38, 148–153. Syldatk, C., May, O., Altenbuchner, J., Mattes, R. and Siemann, M. (1999) Applied Microbiology and Biotechnology, 51, 293–309. May, O., Nguyen, P.T. and Arnold, F.H. (2000) Nature Biotechnology, 18, 317–320. DeSantis, G., Zhu, Z., Greenberg, W.A., Wong, K., Chaplin, J., Hanson, S.R., Farwell, B., Nicholson, L.W., Rand, C.L., Weiner, D.P., Robertson, D.E. and Burk,

57

58

59

60

61

62 63 64

65

66

67

68

M.J. (2002) Journal of the American Chemical Society, 124, 9024–9025. Cedrone, F., Niel, S., Roca, S., Bhatnagar, T., Ait-Abdelkader, N., Torre, C., Krumm, H., Maichele, A., Reetz, M.T. and Baratti, J.C. (2003) Biocatalysis and Biotransformation, 21, 357–364. Reetz, M.T., Torre, C., Eipper, A., Lohmer, R., Hermes, M., Brunner, B., Maichele, A., Bocola, M., Arand, M., Cronin, A., Genzel, Y., Archelas, A. and Furstoss, R. (2004) Organic Letters, 6, 177–180. van Loo, B., Spelberg, J.H.L., Kingma, J., Sonke, T., Wubbolts, M.G. and Janssen, D.B. (2004) Chemistry and Biology, 11, 981–990. Archelas, A. and Furstoss, R. (2001) Current Opinion in Chemical Biology, 5, 112–119. Zou, J.Y., Hallberg, B.M., Bergfors, T., Oesch, F., Arand, M., Mowbray, S.L. and Jones, T.A. (2000) Structure (London), 8, 111–122. Reetz, M.T., Kahakeaw, D. and Lohmer, R. unpublished. Jacobsen, E.N. (2000) Accounts of Chemical Research, 33, 421–431. Benschop, H.P. and de Jong, L.P.A. (1988) Accounts of Chemical Research, 21, 368–374. Wu, F., Li, W.-S., Chen-Goodspeed, M., Sogorb, M.A. and Raushel, F.M. (2000) Journal of the American Chemical Society, 122, 10206–10207. (a) Chen-Goodspeed, M., Sogorb, M.A., Wu, F., Hong, S.-B. and Raushel, F.M. (2001) Biochemistry, 40, 1325–1331. (b) Chen-Goodspeed, M., Sogorb, M.A., Wu, F. and Raushel, F.M. (2001) Biochemistry, 40, 1332–1339. (a) Cunningham, B.C. and Wells, J.A. (1989) Science, 244, 1081–1085. (b) Morrison, K.L. and Weiss, G.A. (2001) Current Opinion in Chemical Biology, 5, 302–307. Hill, C.M., Li, W.-S., Thoden, J.B., Holden, H.M. and Raushel, F.M. (2003) Journal of the American Chemical Society, 125, 8990–8991.

References 69 (a) Rozzell, J.D. and Bommarius, A.S. (2002) In Enzyme Catalysis in Organic Synthesis: A Comprehensive Handbook2nd edn. (eds K. Drauz and H. Waldmann), Wiley-VCH Verlag GmbH, Weinheim, Vol. II, pp. 873–893. (b) Li, T., Kootstra, A.B. and Fotheringham, I.G. (2002) Organic Process Research and Development, 6, 533–538. 70 Matcham, G.W. and Bowen, A.R.S. (1996) Chimica Oggi-Chemistry Today, 14 (6), 20–24. 71 Matcham, G., Bhatia, M., Lang, W., Lewis, C., Nelson, R., Wang, A. and Wu, W. (1999) Chimia, 53, 584–589. 72 Yun, H., Hwang, B.-Y., Lee, J.-H. and Kim, B.-G. (2005) Applied and Environment Microbiology, 4220–4224. 73 (a) Fessner, W.-D. and Walter, C. (1996) Topics in Current Chemistry, 184, 97–194. (b) Silvestri, M.G., Desantis, G., Mitchell, M. and Wong, C.-H. (2003) Topics in Stereochemistry (ed. S.E. Denmark), John Wiley & Sons, Ltd, Weinheim, Vol. 23, pp. 267–342. 74 Fong, S., Machajewski, T.D., Mak, C.C. and Wong, C.-H. (2000) Chemistry and Biology, 7, 873–883. 75 Shelton, M.C., Cotterill, I.C., Novak, S.T.A., Poonawala, R.M., Sudarshan, S. and Toone, E.J. (1996) Journal of the American Chemical Society, 118, 2117–2125. 76 Wada, M., Hsu, C.-C., Franke, D., Mitchell, M., Heine, A., Wilson, I. and Wong, C.-H. (2003) Bioorganic and Medicinal Chemistry, 11, 2091–2098. 77 (a) Hsu, C.-C., Hong, Z., Wasa, M., Franke, D. and Wong, C.-H. (2005) Proceedings of the National Academy of Sciences of the United States of America, 102, 9122–9126. (b) Jennewein, S., Sch€ urmann, M., Wolberg, M., Hilker, I., Luiten, R., Wubbolts, M. and Mink, D. (2006) Biotechnology Journal, 1, 537–548. 78 Williams, G.J., Domann, S., Nelson, A. and Berry, A. (2003) Proceedings of the

79

80

81 82

83

84

85

86

87

88

National Academy of Sciences of the United States of America, 100, 3143–3148. Cornils, B. and Herrmann, W.A. (eds) (1996) Applied Homogeneous Catalysis with Organometallic Compounds, Wiley-VCH Verlag GmbH, Vol. 1–2. Holland, H.L. and Weber, H.K. (2000) Current Opinion in Biotechnology, 11, 547–553. Krow, G.R. (1993) Organic Reaction (New York), 43, 251–798. (a) Bolm, C., Schlingloff, G. and Weickhardt, K. (1994) Angewandte Chemie, 106, 1944–1946; (1994) Angewandte Chemie International Edition in English, 33, 1848–1849. (b) Bolm, C., Frison, J.C., Zhang, Y. and Wulff, W.D. (2004) Synlett, 1619–1621. (c) Bolm, C., Palazzi, C. and Beckmann, O. (2004) In Transition Metals for Organic Chemistry: Building Blocks and Fine Chemicals, 2nd edn. (eds M. Beller and C. Bolm), WileyVCH Verlag GmbH, Weinheim, Vol. 2, pp. 267–274. Strukul, G. (1998) Angewandte Chemie, 110, 1256–1267. (1998) Angewandte Chemie International Edition, 37, 1198–1209. Taschner, M.J. and Black, D.J. (1988) Journal of the American Chemical Society, 110, 6892–6893. Flitsch, S. and Grogan, G. (2002) Enzyme Catalysis in Organic Synthesis (eds K. Drauz and H Waldmann), Wiley-VCH Verlag GmbH, Weinheim, Vol. 2, pp. 1202–1245. Stewart, J.D., Reed, K.W., Zhu, J., Chen, G. and Kayser, M.M. (1996) Journal of Organic Chemistry, 61, 7652–7653. (a) Mihovilovic, M.D., Chen, G., Wang, S., Kyte, B., Rochon, F., Kayser, M.M. and Stewart, J.D. (2001) Journal of Organic Chemistry, 66, 733–738. (b) Mihovilovic, M.D., M€ uller, B., Schulze, A., Stanetty, P. and Kayser, M.M. (2003) European Journal of Organic Chemistry, 2243–2249. (a) Donoghue, N.A., Norris, D.B. and Trudgill, P.W. (1976) European Journal of Biochemistry, 63, 175–192. (b) Walsh, C.T.

j61

j 2 Directed Evolution as a Means to Engineer Enantioselective Enzymes

62

89

90

91

92

93

94

95 96

97

98

and Chen, Y.-C.J. (1988) Angewandte Chemie, 100, 342–352 (1988) Angewandte Chemie International Edition in English, 27, 333–343. (c) Sheng, D., Ballou, D.P. and Massey, V. (2001) Biochemistry, 40, 11156–11167. Reetz, M.T., Brunner, B., Schneider, T., Schulz, F., Clouthier, C.M. and Kayser, M.M. (2004) Angewandte Chemie, 116, 4167–4170 (2004) Angewandte Chemie International Edition, 43, 4075–4078. Mihovilovic, M.D., Rudroff, F., Winninger, A., Schneider, T., Schulz, F. and Reetz, M.T. (2006) Organic Letters, 8, 1221–1224. Malito, E., Alﬁeri, A., Fraaije, M.W. and Mattevi, A. (2004) Proceedings of the National Academy of Sciences of the United States of America, 101, 13157–13162. Fraaije, M.W., Wu, J., Heuts, D.P.H.M., van Hellemond, E.W., Spelberg, J.H.L. and Janssen, D.B. (2005) Applied Microbiology and Biotechnology, 66, 393–400. Bocola, M., Schulz, F., Leca, F., Vogel, A., Fraaije, M.W. and Reetz, M.T. (2005) Advanced Synthesis and Catalysis, 347, 979–986. Schulz, F., Leca, F., Hollmann, F. and Reetz, M.T. (2005) Beilstein Journal of Organic Chemistry, 1, 10. Hollmann, F., Taglieber, A., Schulz, F. and Reetz, M.T. in preparation. (a) Grifﬁn, M. and Trudgill, P.W. (1972) Biochemical Journal, 129, 595–603. (b) Adger, B., Bes, M.T., Grogan, G., McCague, R., Pedragosa-Moreau, S., Roberts, S.M., Villa, R., Wan, P.W.H. and Willetts, A.J. (1997) Bioorganic and Medicinal Chemistry, 5, 253–261. (c) Mihovilovic, M.D., M€ uller, B., Kayser, M.M., Stewart, J.D. and Stanetty, P. (2002) Synlett, 703–706. Alphand, V., Gaggero, N., Colonna, S., Pasta, P. and Furstoss, R. (1997) Tetrahedron, 53, 9695–9706. Dostert, P.L., Benedetti, M.S. and Tipton, K.F. (1989) Medicinal Chemistry Research, 9, 45–89.

99 Beard, T.M. and Turner, N.J. (2002) Chemical Communications (Cambridge U.K.), 246–247. 100 Alexeeva, M., Enright, A., Dawson, M.J. and Mahmoudian, M., Turner, N.J. (2002) Angewandte Chemie, 114, 3309–3312; (2002) Angewandte Chemie International Edition, 41, 3177–3180. 101 Carr, R., Alexeeva, M., Enright, A., Eve, T.S.C., Dawson, M.J. and Turner, N.J. (2003) Angewandte Chemie, 115, 4955–4958 (2003) Angewandte Chemie International Edition, 42, 4807–4810. 102 Dunsmore, C.J., Carr, R., Fleming, T. and Turner, N.J. (2006) Journal of the American Chemical Society, 128, 2224–2225. 103 (a) Sono, M., Roach, M.P., Coulter, E.D. and Dawson, J.H. (1996) Chemical Reviews, 96, 2841–2888. (b) Urlacher, V. and Schmid, R.D. (2002) Current Opinion in Biotechnology, 13, 557–564. (c) Meunier, B., de Visser, S.P. and Shaik, S. (2004) Chemical Reviews, 104, 3947–3980. 104 Holland, H.L. (1998) Biotechnology, 2nd edn. (eds H.-J. Rehm, G Reed, A. P€ uhler and P. Stadler), Wiley-VCH Verlag GmbH, Weinheim, Vol. 8a, pp. 475–533. 105 (a) Farinas, E.T., Schwaneberg, U., Glieder, A. and Arnold, F.H. (2001) Advanced Synthesis and Catalysis, 343, 601–606. (b) Glieder, A., Farinas, E.T. and Arnold, F.H. (2002) Nature Biotechnology, 20, 1135–1139. (c) Peters, M.W., Meinhold, P., Glieder, A. and Arnold, F.H. (2003) Journal of the American Chemical Society, 125, 13442–13450. 106 M€ unzer, D.F., Meinhold, P., Peters, M.W., Feichtenhofer, S., Griengl, H., Arnold, F.H., Glieder, A. and de Raadt, A. (2005) Chemical Communications (Cambridge U.K.), 2597–2599. 107 Landwehr, M., Hochrein, L., Otey, C.R., Kasrayan, A., B€ackvall, J.-E. and Arnold, F.H. (2006) Journal of the American Chemical Society, 128, 6058–6059. 108 (a) Kubo, T., Peters, M.W., Meinhold, P. and Arnold, F.H. (2006) Chemistry-A European Journal, 12, 1216–1220. (b) Budde, M., Morr, M., Schmid, R.D. and

References Urlacher, V.B. (2006) Chembiochem: A European Journal of Chemical Biology, 7, 789–794. 109 (a) Hancock, S.M., Vaughan, M.D. and Withers, S.G. (2006) Current Opinion in Chemical Biology, 10, 509–519. (b) Dion, M., Nisole, A., Spangenberg, P., Andre, C., Glottin-Fleury, A., Mattes, R., Tellier, C. and Rabiller, C. (2001) Glycoconjugate Journal, 18, 215–223. 110 (a) Vedha-Peters, K., Gunawardana, M., Rozzell, J.D. and Novick, S.J. (2006) Journal of the American Chemical Society, 128, 10923–10929. (b) Baik, S.-H., Ide, T., Yoshida, H., Kagami, O. and Harayama, S. (2003) Applied Microbiology and Biotechnology, 61, 329–335. (c) Morikawa, S., Nakai, T., Yasohara, Y., Nanba, H., Kizaki, N. and Hasegawa, J. (2005) Bioscience Biotechnology and Biochemistry, 69, 544–552. (d) Bali, S., OHare, H.M. and Weissman, K.J. (2006) Chembiochem: A European Journal of Chemical Biology, 7, 478–484. 111 (a) James, L.C. and Tawﬁk, D.S. (2003) Trends in Biochemical Sciences, 28, 361–368. (b) Bornscheuer, U.T. and Kazlauskas, R.J. (2004) Angewandte Chemie, 116, 6156–6165; (2004) Angewandte Chemie International Edition, 43, 6032–6040. (c) Aharoni, A., Gaidukov, L., Khersonsky, O., McQ Gould, S., Roodveldt, C. and Tawﬁk, D.S. (2005) Nature Genetics, 37, 73–76.

112 See for example: (a)Funke, S.A., Otte, N., Eggert, T., Bocola, M., Jaeger, K.-E. and Thiel, W. (2005) Protein Engineering, Design and Selection, 18, 509–514. (b) Kato, R., Nakano, H., Konishi, H., Kato, K., Koga, Y., Yamane, T., Kobayashi, T. and Honda, H. (2005) Journal of Molecular Biology, 351, 683–692. (c) Moore, G.L. and Maranas, C.D. (2004) AIChE Journal, 50, 262–272. (d) Weinreich, D.M., Delaney, N.F., DePristo, M.A. and Hartl, D.L. (2006) Science, 312, 111–114. (e) Fox, R., Roy, A., Govindarajan, S., Minshull, J., Gustafsson, C., Jones, J.T. and Emig, R. (2003) Protein Engineering, 16, 589–597. (f) Nakahara, T., Hindsgaul, O., Palcic, M.M. and Nishimura, S.-I. (2006) Protein Engineering, Design and Selection, 19, 571–578. (g) DiTursi, M.K., Kwon, S.-J., Reeder, P.J. and Dordick, J.S. (2006) Protein Engineering, Design and Selection, 19, 517–524. (h) Bogarad, L.D. and Deem, M.W. (1999) Proceedings of the National Academy of Sciences of the United States of America, 96, 2591–2595 (i) Mayer, M.M., Hochrein, L. and Arnold, F.H. (2006) Protein Engineering, Design and Selection, 19, 563–570. 113 (a) Reetz, M.T., Rentzsch, M., Pletsch, A. and Maywald, M. (2002) Chimia, 56, 721–723. (b) Reetz, M.T. (2001) patent application DE-A 101 29 187.6 (Optimization of Synthetic Catalysts by Means of Directed Evolution).

j63

j65

3 The Search for New Enzymes Jean-Louis Reymond and Wolfgang Streit

3.1 Introduction

Other than the opportunity offered by optimizing a biocatalytic process by engineering and modifying existing enzymes by directed evolution methods [1], one can look for an entirely new enzyme to solve the problem at hand. This option might open unforeseen advantages and allow a total redesign of a process if a new reaction type is found. It might also be necessary if an intellectual property limitation is encountered in a given process. In principle, there are two options to search for a new enzyme: (i) design a new enzyme starting with the knowledge of a reaction mechanism; and (ii) the search for a yet undiscovered enzyme in biodiversity, in particular, through mining the metagenome. Overall, these two approaches represent radically different points of view to the same challenge, yet follow similar workﬂows via recombinant or synthetic libraries and high-throughput screening for function (Figure 3.1) [2]. In the enzyme design approach, as discussed in the ﬁrst part of this chapter, one attempts to utilize the mechanistic understanding of chemical reactions and enzyme structure to create a new catalyst. This approach represents a largely academic research ﬁeld aiming at fundamental understanding of biocatalysis. Indeed, the invention of functional artiﬁcial enzymes can be considered to be the ultimate test for any theory on enzyme mechanisms. Most artiﬁcial enzymes, to date, do not fulﬁll the conditions of catalytic efﬁciency and price per unit necessary for industrial applications. The biodiversity mining approach focuses on expressing possible enzymes found in microbial collections and metagenomic expression libraries [3]. In contrast to the rather academic setting of the enzyme design approach, the screening of microbial collections and the metagenome is integrated in both academia and industry, because it represents a very practical method to solve a catalysis problem whenever a new enzyme is necessary. The approach is particularly successful when new enzymes are searched to carry out a known reaction type on a new type of substrate, or with a different stereoselectivity compared to already known enzymes. In such cases,

Asymmetric Organic Synthesis with Enzymes. Edited by Vicente Gotor, Ignacio Alfonso, and Eduardo Garcıa-Urdiales Copyright Ó 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 978-3-527-31825-4

j 3 The Search for New Enzymes

66

Synthetic libraries Reaction mechanisms Recombinant libraries

HTS assays

New enzymes

Biodiversity Figure 3.1 The search for new enzymes.

screening enzyme collections such as those offered by a range of enzyme companies is a practical ﬁrst step. In this chapter, we focus particularly on the metagenome because it represents a generalization of biodiversity mining by extending its boundary to all life forms, including in particular, uncultivable microorganisms.

3.2 Mechanism-based Enzyme Design

The mechanism of any given reaction provides information about the catalytic groups and nature of the transition state involved [4]. This information can be implemented into a structural framework such as a protein or peptide, which is either expressed in recombinant format or prepared by chemical synthesis. The approach includes (i) transition state mimicry-based designs such as catalytic antibodies [5] and the related computational redesign of binding pockets around a transition state based on de novo protein design algorithms [6,7]; (ii) the rational design of new catalysts on protein and enzyme basis, either by design [8], searching for promiscuous activities in existing enzymes by substrate and/or active site modiﬁcations [9], by cofactor engineering [8,10], or by installing a catalytic metal complex into a noncatalytic protein [11]; and (iii) the search for catalytic activity in synthetic nonprotein macromolecules such as synthetic peptides [12] and dendrimers [13]. 3.2.1 Catalytic Antibodies

Catalytic antibodies were ﬁrst reported by Lerner and Schultz in the 1980s [5]. Catalytic antibodies are obtained by immunizing a laboratory animal using a stable transition state analog (TSA) of a chemical reaction as hapten. Following the basic principle that chemical catalysis is equivalent to binding to the transition state of a chemical reaction, one can expect that an antibody (Ab) forming a complex with the TSA for a reaction from S to P, with a dissociation constant Ki, might form a similarly tight complex with the real transition state of the reaction (TS ), with a dissociation constant KTS, and therefore provide rate acceleration for the reaction (G #) (Figure 3.2) [14]. TSA design considers the charge distribution and shape of the transition state

3.2 Mechanism-based Enzyme Design

TS* ∆∆G AbTS* S

P

K TS

Ki Ab

AbTS*

AbTSA TSA

TS*

Figure 3.2 Transition state analog principle for the design of catalytic antibodies.

along the same lines used to design TSA enzyme inhibitors. Covalent immunization using a reactive hapten to trap a catalytic side chain on the antibody, which borrows from the concept of suicide inhibitors of enzymes, has also been used with success [15]. Catalytic antibodies are particularly interesting, because they allow one to tackle reaction types not encountered in natural enzymes. One such example is the enantioselective protonation of enol ether (1) to form a single enantiomer of ketone (2) (Figure 3.3). This reaction was found to be catalyzed efﬁciently and with complete enantioselectivity in favor of the (S) enantiomer by antibody 14D9, a monoclonal antibody produced by a hybridoma cell line resulting from an immunization using hapten (3) as antigen [16]. The hapten design and reaction mechanism of antibody 14D9 is representative of the catalytic antibody approach to new enzymes. Hapten (1) combines a strongly immunogenic aromatic group that can induce substrate recognition with a quaternary ammonium group to induce a hydrophobic binding pocket equipped with a carboxylic acid to be used as catalytic functional Ab 14D9 pH 6, 20 °C MeO

O H

kcat = 0.4 s –1; kcat /kuncat = 104

Ar

Ar

1

(S)-2 Carrier protein Me

N

> 98 % ee

= Ar H N

OH

O 3 Figure 3.3 Antibody 14D9 catalyzes the enantioselective protonation of enol ethers.

j67

j 3 The Search for New Enzymes

68

group for the reaction. The enantioselective protonation of enol ether (1) by antibody 14D9 tolerates a variety of substrates [17], and is sufﬁciently high to allow preparative scale reactions on the gram scale [18]. The molecular mechanism of the enantioselective protonation reaction by antibody 14D9 was revealed by a crystal structure analysis [19]. A catalytic carboxyl group AspH101 was found at the bottom of the catalytic pocket and found to be necessary for catalysis by mutagenesis to Asn or Ala. The mechanism or protonation involves an overall syn addition of water to the enol ether in a chiral binding pocket ensuring complete enantioselectivity (Figure 3.4). Like many other antibodies, the activity of antibody 14D9 is sufﬁcient for preparative application, yet it remains modest when compared to that of enzymes. The protein is relatively difﬁcult to produce, although a recombinant format as a fusion with the NusA protein was found to provide the antibody in soluble form with good activity [20]. It should be mentioned that aldolase catalytic antibodies operating by an enamine mechanism, obtained by the principle of reactive immunization mentioned above [15], represent another example of enantioselective antibodies, which have proven to be preparatively useful in organic synthesis [21]. One such aldolase antibody, antibody 38C2, is commercially available and provides a useful alternative to natural aldolases to prepare a variety of enantiomerically pure aldol products, which are otherwise difﬁcult to prepare, allowing applications in natural product synthesis [22]. Ar

Ar

H O H O

TyrH35

H O H O H O H O

O H O

TyrH35

AspH101

Ar H

TyrH35

AspH101 TyrL36

TyrL36

O H O

O H O H O H O

H O

O O H

H O

AspH101 TyrL36

Figure 3.4 Protonation mechanism for 14D9.

3.2 Mechanism-based Enzyme Design

3.2.2 Rational Design of New Catalysts on Enzyme and Protein Basis

One of the most direct questions to ask in the perspective of enzyme design is whether an already existing protein with a binding pocket might be turned into a new catalyst by introducing catalytic residues directly, rather than by the elaborated TSA mimicry approach used for catalytic antibodies, hoping to create a new biocatalyst that could harness both the activity and the selectivity, in particular stereoselectivity, that is possible with enzymes. In principle, numerous reports have detailed the possibility to modify an enzyme to carry out a different type of reaction than that of its attributed function, and the possibility to modify the cofactor of the enzyme has been well explored [8,10]. Recently, the possibility to directly observe reactions, normally not catalyzed by an enzyme when choosing a modiﬁed substrate, has been reported under the concept of catalytic promiscuity [9], a phenomenon that is believed to be involved in the appearance of new enzyme functions during the course of evolution [23]. A recent example of catalytic promiscuity of possible interest for novel biotransformations concerns the discovery that mutation of the nucleophilic serine residue in the active site of Candida antarctica lipase B produces a mutant (Ser105Ala) capable of efﬁciently catalyzing the Michael addition of acetyl acetone to methyl vinyl ketone [24]. The oxyanion hole is believed to be complex and activate the carbonyl group of the electrophile, while the histidine nucleophile takes care of generating the acetyl acetonate anion by deprotonation of the carbon (Figure 3.5).

O

O

Gln106 O O N H H N H O

CAL-B Ser105Ala

O

Thr40

+ O

O

O O H N

Asp187

N H

O

His224

O

O

O Figure 3.5 Michael addition catalyzed by the Ser105Ala C. antarctica lipase B mutant.

j69

j 3 The Search for New Enzymes

70

O

BSA, Cu(I)-phthalocyanin

N

H H

+ Formate buffer pH 4

O N 77% yield 92% de exo 93% ee

Figure 3.6 Enantioselective Diels–Alder reaction catalyzed by BSA and a Cu-phthalocyanin [26].

It has also been observed that noncatalytic proteins can be brought to exhibit unusual catalytic properties. Serum albumins such as bovine serium albumin (BSA), for instance, promote a range of reactions such as the base-catalyzed decomposition of benzisoxazole, which has been studied in great detail [25]. BSA possesses a binding pocket for large hydrophobic aromatic compounds and provides a chiral environment suitable for other types of catalysis as well, in particular, for porphyrin-type metal complexes for a range of metal-catalyzed processes such as reductions, oxidations, and Diels–Alder cycloadditions, as recently reviewed [11b]. One such recent example is the combination of copper phthalocyanine with BSA to produce a new biocatalyst catalyzing a highly enantioselective Diels–Alder cycloaddition reaction [26]. Interestingly, a similar highly enantioselective system has also been reported combining the copper catalyst with DNA as a source of chirality [27]. These examples are part of a broader design scheme to combine catalytic metal complexes with a protein as chiral scaffold to obtain a hybrid catalyst combining the catalytic potential of the metal complex with the enantioselectivity and evolvability of the protein host [11]. One of the ﬁrst examples of such systems combined a biotinylated rhodium complex with avidin to obtain an enantioselective hydrogenation catalyst [28]. Most signiﬁcantly, it has been shown that mutation-based improvements of enantioselectivity are possible in these hybrid catalysts as for enzymes (Figure 3.7) [29]. Streptavidin 5 atm H 2

CO2H

CO2H

NHAc

NHAc Quantity yield 94% ee (R)

O HN

P

NH S

N

RhH2 P

O

Figure 3.7 The biotin–streptavidin system for enantioselective hydrogenation.

3.3 Metagenomics

3.2.3 Synthetic Enzyme Models

The ﬁeld of synthetic enzyme models encompasses attempts to prepare enzymelike functional macromolecules by chemical synthesis [30]. One particularly relevant approach to such enzyme mimics concerns dendrimers, which are treelike synthetic macromolecules with a globular shape similar to a folded protein, and useful in a range of applications including catalysis [31]. Peptide dendrimers, which, like proteins, are composed of amino acids, are particularly well suited as mimics for proteins and enzymes [32]. These dendrimers can be prepared using combinatorial chemistry methods on solid support [33], similar to those used in the context of catalyst and ligand discovery programs in chemistry [34]. Peptide dendrimers used multivalency effects at the dendrimer surface to trigger cooperativity between amino acids, as has been observed in various esterase enzyme models [35]. An interesting case in the perspective of artiﬁcial enzymes for enantioselective synthesis is the recently described peptide dendrimer aldolases [36]. These dendrimers utilize the enamine type I aldolase mechanism, which is found in natural aldolases [37] and antibodies [21]. These aldolase dendrimers, for example, L2D1, have multiple N-terminal proline residues as found in catalytic aldolase peptides [38], and display catalytic activity in aqueous medium under conditions where the small molecule catalysts are inactive (Figure 3.8). As most enzyme models, these dendrimers remain very far from natural enzymes in terms of both activity and selectivity, and at present should only be considered in the perspective of fundamental studies.

3.3 Metagenomics

Metagenomics has dramatically increased the speed of discovery of novel biocatalysts by enabling scientists to tap directly into the entire diversity of enzymes held within natural microbial populations. The genomes of as yet uncultured microbes represent a shear unlimited source for novel genes and chemical compounds, to be used by the biotechnology industries for the development of novel products [3]. The dilemma of being unable to cultivate the vast majority of the microorganisms present in most microbial niches has forced scientists to develop methods to exploit the genomes of these bacteria for biotechnology without prior cultivation. The relatively young ﬁeld of metagenomics developed from advances made in DNA extraction and cloning from environmental samples [39,40]. The key technologies have been outlined in a landmark publication in 1991. This development coincided with the observation that prokaryotic diversity had been greatly underestimated and it became apparent that there remained a great deal of biocatalytic potential locked within the unculturable majority of prokaryotes [z3,41a]. Current computed estimates of soil diversity are in the range of 1 million species per 1 g of soil [42]. Metagenomics now enables access to the biocatalytic potential of unculturable bacteria [43]. Metagenomics pursues two lines of activities: ﬁrst, it uncovers novel enzymes and molecules for biotechnological

j71

j 3 The Search for New Enzymes

72

O

OH O

Dendrimer L2D1 (1 mol%) H

O2 N

O2N 61% ee HN

+

NH NH

O O

H 3N O NH

O

O H N

N

+

H N

HN

N H

O

+

O

O

O

NH

H 3N

H N

NH 3+ +

NH

NH

H N

O NH

NH O

NH 2 O

O

NH HN

H 3N

N H

HN

H N

O

O +

O

OH O N

O

HN

H N

O OHN

NH 3 +

NH

O

NH

O

N

N H

NH

NH3 + N H

O O

O O NH

H3N

HN O

O

O

HN

HN NH3 +

NH 3+

NH N H

NH 3+

O

O O

OH O

N HN

H3N HN

O HN

H N

H N

O

O O

NH 3+

NH L2D1

Figure 3.8 Aldol reaction and peptide dendrimer catalyst.

and pharmaceutical applications; and secondly, it generates new knowledge on the microbial ecology of the studied niches. In this review, we discuss only the biotechnological aspects of metagenomics that are related to the identiﬁcation of speciﬁc enzymes and genes. 3.3.1 Construction of Metagenome-derived DNA Libraries 3.3.1.1 Selection of the Environment A range of novel enzyme detection strategies have been developed, which target different stages in the process of metagenomic library construction and screening. The success of a metagenome-based search for enzymes depends, in general, on the

3.3 Metagenomics

selection of the environment. In general, the published studies can be grouped into three categories. The ﬁrst group selects environments naturally enriched for the target biocatalyst such as the search for xylanases in insect gut [44]. The second group selects highly genomically diverse environments such as soil and either directly extracts and clones the DNA or subjects the community ﬁrst to enrichment and then proceeds with extraction and cloning. The third group targets extreme environments and searches for biocatalysts that are stable under the conditions experienced by the microorganisms in the environment [45]. 3.3.1.2 Cloning Strategies Other than the careful selection of the environment, the cloning strategy is also of considerable importance. The overall cloning strategy of metagenome-derived DNA fragments is outlined in Figure 3.9. In general, the metagenomic DNA is cloned into either large insert or small insert libraries. Large insert libraries can be generated in cosmids [46,47], bacterial artiﬁcial chromosomes (BACs)[48], or fosmids [45,49]. Small insert libraries are maintained in plasmid vectors including pBluescript SKþ [50], pUC19 [51], pZero-2 [52], and ZAP phagemid vector [45]. In the case of pZero-2, a lethal gene is inactivated by the incorporation of environmental DNA, which keeps the level of clones containing a self-ligated vector to a minimum [52]. Small insert libraries contain on average ten times more clones than large insert libraries covering the same amount of environmental DNA. Although the reduced clone number makes large insert libraries easier to screen, the larger insert size and lower copy number makes detection of weakly expressed foreign genes more difﬁcult [42b]. 3.3.1.3 Screening and Detection Technologies With respect to the technologies used for screening, two approaches are used: functional searches and sequence-based searches. Sequence-based searches are always more conservative, but will certainly beneﬁt from the vast amount of data produced by the environmental sequencing projects [53]. However, functional screening strategies aim to detect the biocatalytic reaction ﬁrst and then characterize the gene responsible. In this way completely novel enzymes can be discovered. This approach requires that the metagenomic DNA be expressed in a heterologous host, usually Escherichia coli. As can be expected with an expression-based system, it is limited when factors required for transcription and translation of the metagenomic DNA are not present in the host. A survey of 32 prokaryotic genomes found that only 40% of genes could be expressed using E. coli [45]. Also, even when a gene is expressed, the level of expression can be so low that it cannot be detected using functional screening. Recently, a transposon that enables inducible expression over T7 promoters in both directions was developed [54]. This transposon could be used to enhance gene expression and increase the likelihood of ﬁnding metagenomic enzymes through functional screening. The heterologous hosts used in metagenomics are E. coli, Pseudomonas putida, Streptomyces lividans, and Rhizobium leguminosarum [55,56]. In each of these studies, it was clearly demonstrated that the expression of metagenomic protein is limited owing to a limited number of hosts in most laboratories. Finally, it is noteworthy that the detection systems that are used

j73

j 3 The Search for New Enzymes

74

Figure 3.9 Cloning and DNA–isolation strategies commonly used in metagenomics.

are also of importance for the frequency of gene detection. A very promising system is the SIGEX system. In the SIGEX system, the environmental DNA is cloned into a vector containing a gfp reporter [57]. Clones carrying the catabolic genes are activated in the presence of the target substrate. When the catabolite genes are expressed, they in turn activate the gfp reporter causing the cell to ﬂuoresce. This, in combination with ﬂuorescence activating cell sorting (FACS), greatly increases the number of recombinant clones that can be screened [57]. 3.3.1.4 Major Problems that Need to be Addressed The rapidly increasing number of novel metagenome-derived genes also led to the identiﬁcation of major bottlenecks. These are mainly linked to the efﬁciency of heterologous gene expression and the amount of protein produced, in particular,

3.3 Metagenomics

when subsequent biotechnological applications are envisaged. Therefore, a need exists to develop novel experimental systems for high-throughput expression of metagenomic genes and puriﬁcation of the respective proteins. Additional novel expression host strains must be developed, which should include both prokaryotic and eukaryotic organisms. In any case, the enormous potential of metagenomics for modern biotechnology is obvious, but further and quick advances of technologies are necessary to fully exploit the genetic information of nonculturable microbes as a virtually unlimited resource for biotechnology. Among these technologies, it is mostly the screening that limits the identiﬁcation of novel biocatalysts. The urgent need to develop screening technologies for the reactions that have been unusual and novel will hopefully help metagenomics to become an even more important key technology in modern biotechnology. 3.3.2 The Genomes of Not Yet Cultured Microbes as Resources for Novel Genes

Metagenomic libraries have been screened for a wide range of biocatalysts [3]. However, many of the published and metagenome-derived enzymes/genes have not been characterized to such an extent that would allow their immediate use in biotechnological applications. In the following section, we highlight examples of the successful isolation enzymes and other valuable biomolecules from metagenomes. 3.3.2.1 Polysaccharide Degrading/Modifying Enzymes Agarases are the enzymes that can liquefy agar, and they can be divided into and agarases, depending on whether they cleave the -L-(1,3) or the -D-(1,4) linkages of the polymer. Although agarases are normally only found in marine microbes, the screening of a soil metagenomic library identiﬁed a total of four agarolytic clones containing 12 agarase genes [58]. This highlights the potential of using metagenomics to investigate environments for novel biocatalysts, which are normally detected in isolates from unrelated environments. Amylases have also been in the focus of a number of metagenome studies. At least ﬁve articles reporting the detection and characterization of novel amylolytic enzymes from metagenomic DNA libraries have been published [58–61]. Of the 14 amylolytic clones only four have been puriﬁed and characterized. One of the characterized amylases, from a soil metagenome, displayed interesting characteristics in that it is stable and active under alkaline conditions, with a pH optimum at pH 9, a characteristic required of amylases in detergents [61,62]. The other three amylases were active under acidic pH and high temperature conditions, and these were then subjected to gene reassembly in order to obtain one enzyme displaying optimal properties [60]. Functional screening of a soil metagenomic library for cellulases revealed a total of eight cellulolytic clones, one of which was puriﬁed and characterized [58]. Despite the fact that this library had been generated from a soil sample collected from a

j75

j 3 The Search for New Enzymes

76

Figure 3.10 Typical screening plate for cellulase and amylase positive clones from enrichment cultures. Clear halos indicate the presence of positive clones and the halos diameters give a first idea about the overall activities of the respective enzymes.

nonextreme environment, the cellulase displayed a high level of stability over a broad pH range, up to pH 9, it was stable at 40 C for up to 11 hours, and was highly halotolerant, being active and stable in 3 M NaCl [58]. Metagenomic screening of extreme environments, soda lakes in Africa and Egypt, detected more than a dozen cellulases, some of which displayed habitat-related halotolerant characteristics [63]. One of the earliest articles presenting metagenome-derived biocatalysts reported the detection of cellulases from a thermophilic, anaerobic digester fueled by lignocellulose [64]. While most metagenomic surveys for novel cellulases concentrate on extreme environments, there is sufﬁcient evidence that nonextreme, and therefore highly genetically diverse, environments also contain a range of cellulases that are highly stable and suitable for industrial applications [65]. Chitinases are responsible for the breakdown and recycling of chitin. With several gigatons of chitin being produced annually, it is the second most abundant polymer

3.3 Metagenomics

in nature after cellulose [66]. Chitinases have several biotechnological applications including increasing plant resistance to fungal disease [67]. To date, there has been only a few metagenomic studies targeting chitinases from marine environments [68]. The identiﬁed enzymes need to be characterized on a more detailed level before application in biotechnological processes. Complete hydrolysis of xylan to xylose involves the activity of xylosidases. There have been three studies detecting xylanases/xylosidases in the metagenomes of diverse environments: an insect gut, a thermophilic, anaerobic digester, and waste lagoon of a dairy farm [44,64]. Screening of the lagoon resulted in the detection and characterization of one xylanase, which displayed habitat-related properties in that it was most active at lower temperatures [69]. The four xylosidases detected in the digester were not investigated further, but the cellulases found in the same study were thermophilic [64]. The metagenomic survey of the insect gut discovered four xylanases, which were phylogenetically distant from all other known xylanases, suggesting that they had evolved independently [44]. These xylanases are novel and produce unique hydrolysis products. 3.3.2.2 Lipolytic Biocatalysts The development of metagenomics has greatly accelerated the discovery of novel biocatalysts, many displaying unusual properties. To date, at least 76 esterase or lipase positive clones derived from metagenomes have been reported (Table 3.1). The level of characterization of these novel lipolytic genes ranges from DNA restriction and sequencing analysis to determine clone diversity [48,58] to detailed biochemical analysis of the puriﬁed enzyme [45,75]. Of the 76 lipolytic positive clones listed, only 11 have been overexpressed, puriﬁed, and subjected to detailed biochemical characterization. It is because of the high number of novel enzymes, which can be detected by the screening of a single metagenomic library, and the amount of time and effort required to fully characterize one enzyme, which causes this bottleneck. Among the characterized metagenomic esterases found so far are two of the largest esterases known: a 325-kDa esterase from a deep-sea hypersaline anoxic basin, and 336-kDa octameric esterase from a drinking water bioﬁlm [45,74]. The esterases from the deep-sea hypersaline basin display habitat-related properties in that they were most active at alkaline pH and displayed higher activities under high-pressure conditions [75]. The esterases from soil and a drinking water bioﬁlm displayed unusual properties, which could not be related to their environment [75]. They were highly stable at alkaline pH and displayed unique substrate spectra with EstA3 being able to hydrolyze substrates such as 7-[3-octylcarboxy-(3-hydroxy-3-methyl-butyloxy)]coumarin, a normally unreactive secondary ester [3]. 3.3.2.3 Vitamin Biosynthesis Metagenomics has been applied to the search for novel genes encoding the synthesis of vitamins such as biotin and vitamin C [47,76]. Seven cosmids were detected in metagenomic libraries obtained after avidin enrichment of environmental samples. The highest levels of biotin production in this study were detected in a cosmid

j77

j 3 The Search for New Enzymes

78

Table 3.1 Lipolytic genes identified in metagenome screens.

Source

Vector

Total size of DNA screened

Number of positive clones

Number of clones screened

Soil

Plasmid Plasmid Plasmid BAC Cosmid Phagemid Phagemid Fosmid Plasmid Fosmid Fosmid Phagemid

474.5 Mb 1170 Mb 637 Mb 100 Mb 49.8 Mb 360 Mb 600 Mb 1179.5 Mb 114 Mb 40 Mb 408.6 Mb 2000 Gb

1 1 2 2 1 2# 1 8 11 4 10 5

730 000 180 000 49 000 1824 1532 30 000 100 000 4213 2727 200 1021 —

Phagemid Fosmid

77 Mb 210 Mb

11 5

1273 1200

[45] [49]

Cosmid Cosmid Fosmid

56 Mb 87.5 Mb 13.5 Gb

6 1 6

267 2500 64 333

[75]

Soil Soil Kenyan soda lakes Topsoil Pondwater Hot spring Hot spring Deep-sea hypersaline basin Cow gut Mud ﬂats, beach, and forest Drinking water Soil Korean tidal ﬂat

Reference [70] [48] [66] [71] [72] [51] [73] [74] [45]

[69]

obtained from forest soil [47]. In the search for enzymes involved in vitamin C production from glucose, two novel 2,5-diketo-D-gluconic acid reductases were PCR ampliﬁed from the environment [76]. 3.3.2.4 Nitrilases, Nitrile Hydratases, and Amidases Chemical hydrolysis of nitriles requires the presence of strong acids or bases at high temperatures and normally results in low yields. Therefore, there is much interest in using biocatalysts to carry out this reaction instead [77]. Nitrile hydratases are used for the production of acrylamide and the vitamin nicotinamide [78]. In one metagenomic study, involving the screening of 3-Gb DNA, 12 novel nitrile dehydratases were detected [79]. In another general screening of a soil metagenomic library for biocatalysts, one amidase positive clone was detected [58]. Amidases are used in the biosynthesis of ß-lactam antibiotics, and in a study targeting amidases of the soil metagenome using enrichment seven amidase positive clones were detected, one of which encoded a novel penicillin acylase [52,80]. Nitrilases are quite rare in bacterial genomes and less than 20 were reported prior to the application of metagenomics for their detection in environmental DNA [81]. Two studies targeting environmental genomes report the detection of more than 337 novel nitrilases. This has dramatically increased the amount of information about nitrilases, and the newly discovered diversity can be applied for the enantioselective production of hydroxy carboxylic acid derivatives [81].

3.3 Metagenomics

3.3.2.5 Oxidoreductases/Dehydrogenases A metagenomic study searching for the diversity of bacteria in the environment capable of utilizing 4-hydroxybutyrate found ﬁve clones displaying novel 4-hydroxybutyrate dehydrogenase activity [82]. In a recent metagenomic study, the genes involved in metabolism of poly-3-hydroxybutyrate, a compound being considered as a substitute for fossil fuel-derived polymers, were screened for environmental libraries [83]. They found novel short chain dehydrogenases/reductases, which had <35% similarity to known enzymes and thus could not have been detected using hybridization-based techniques such as PCR. Alcohol oxidoreductases capable of oxidizing short chain polyols are useful biocatalysts in industrial production of chiral hydroxy esters, hydroxy acids, amino acids, and alcohols [83]. In a metagenomic study without enrichment, a total of 24 positive clones were obtained and tested for their substrate speciﬁcity. To improve the detection frequency, enrichment was performed using glycerol or 1,2-propanediol and further 24 positive clones were detected in this study. 3.3.2.6 Proteases Proteases are hydrolytic enzymes with important application in industries, in particular, in detergent and in the food industry. A metagenomic study in which 100 000 plasmid clones were screened for proteolytic activity found one positive clone, which was determined to be novel by sequencing analysis [84]. 3.3.2.7 Glycerol Hydratases In industrial processes, 1,3-propanediol is used for the production of polyester ﬁbers, polyurethanes and cyclic compounds [85]. 1,3-Propanediol can be produced from glucose with the limiting step catalyzed by glycerol dehydratase. A metagenomic survey for glycerol hydratases from the environment resulted in seven positive clones, one of which displayed a level of catalytic efﬁciency and stability making it ideal for application in the production of 1,3-propanediol from glucose. 3.3.2.8 Antibiotics and Pharmaceuticals There have been at least two studies reporting the successful screening of soil metagenomic libraries for indirubin, a microbial product used in the treatment of diseases including leukemia [86]. A range of novel antibiotics have been detected in metagenomic libraries [87]. Palmitoylputrescine was detected in metagenomic DNA from a Bromeliad tank [87] and an isocyanide containing antibiotic was found in a soil metagenomic library [87]. Screening of seven different soil metagenomic libraries revealed 11 clones producing long-chain N-acyltyrosine antibiotics and analysis of their syntheses indicated that 10 of them were novel [87]. All the metagenomederived novel antibiotics detected to date have been found in libraries maintained in E. coli hosts except the terragines, which were in a streptomycete host [56]. In addition to these studies, plenty of information has been gained about the diversity of natural antibiotic resistance mechanisms [88]. The search for novel antimicrobial genes and compounds is probably the ﬁeld that is advancing most rapidly in metagenomics.

j79

j 3 The Search for New Enzymes

80

3.4 Conclusion

The search for new enzymes is clearly being driven into two parallel and complementary directions. On the one hand, mechanism-based enzyme design is being pursued to push our fundamental understanding of enzyme catalysis. This ﬁeld provides a driving force in the area of synthetic bioorganic and macromolecular chemistry. As commented in the introduction, most examples of designed artiﬁcial enzymes fall short of natural enzymes in terms of catalytic efﬁciency. One must acknowledge that enzyme design remains a largely unsolved problem, which makes it all the more interesting as a challenge capable of yielding signiﬁcant scientiﬁc advances. These might arise by advancing theoretical understanding of structure–function relationship and catalysis itself, or from developing more efﬁcient experimental protocols for mimicking natural selection/evolution in the laboratory. In parallel, at present, metagenomics embodies a complementary side in the search for new enzymes, where the scientiﬁc quest is not about reﬁned understanding but rather in exploring what nature has to offer. In truth, the revelation of the enormous genetic diversity in the biosphere beyond all that was imagined justiﬁes the expectation that a very large number of discoveries await to be made in this ﬁeld. A signiﬁcant number of metagenome-derived genes have been identiﬁed until now. Future work will have to demonstrate whether these genes or their products can be implemented into viable biotechnological processes.

References 1 (a) Reetz, M.T. (2001) Angewandte Chemie International Edition, 40, 284–310. (b) Bloom, J.D., Meyer, M.M., Meinhold, P., Otey, C.R., MacMillan, D. and Arnold, F.H. (2005) Current Opinion in Structural Biology, 15, 447–452. 2 Reymond, J.-L. (ed.) (2006) Enzyme Assays: High-Throughput Screening, Genetic Selection and Fingerprinting, Wiley-VCH, Weinheim, Germany. 3 (a) Streit, W.R. and Schmitz, R.A. (2004) Current Opinion in Microbiology, 7, 492–498. (b) Streit, W.R., Daniel, R. and Jaeger, K.-E. (2004) Current Opinion in Biotechnology, 15, 285–290. 4 (a) Mader, M.M. and Bartlett, P.A. (1997) Chemical Reviews, 97, 1281–1302. (b) Mulholland, A.J. (2005) Drug Discovery Today, 10, 1393–1402.

5 (a) Lerner, R.A., Benkovic, S.J. and Schultz, P.G. (1991) Science, 252, 659–667. (b) Schultz, P.G. and Lerner, R.A. (1993) Accounts of Chemical Research, 26, 391–395. (c) Schultz, P.G. and Lerner, R.A. (1995) Science, 269, 1835–1842. (d) MacBeath, G. and Hilvert, D. (1996) Chemistry and Biology, 3, 433–445. (e) Reymond, J.-L. (1999) Topics in Current Chemistry, 200, 59. (f) Thomas, N.R. (1996) Natural Product Reports, 13, 479–511. (g) Stevenson, J.D. and Thomas, N.R. (2000) Natural Product Reports, 17, 535–537. 6 Dwyer, M.A., Looger, L.L. and Hellinga, H.W. (2004) Science, 304, 1967–1971. 7 (a) Bolon, D.N., Voigt, C.A. and Mayo, S.L. (2002) Current Opinion in Chemical Biology, 6, 125–129. (b) Kuhlman, B. and Baker, D. (2004) Current Opinion in Structural

References

8

9 10

11

12

13

14

15 16

17

18

19

Biology, 14, 89–95. (c) Dwyer, M.A. and Hellinga, H.W. (2004) Current Opinion in Structural Biology, 14, 495–504. (d) Moffet, D.A. and Hecht, M.H. (2001) Chemical Reviews, 101, 3191–3203. (a) Qi, D., Tann, C.-M., Haring, D. and Distefano, M.D. (2001) Chemical Reviews, 101, 3081–3112. (b) Andersson, L.K., Caspersson, M. and Baltzer, L. (2002) Chembiochem: A European Journal of Chemical Biology, 8, 3687–3689. Hult, K. and Berglund, P. (2003) Current Opinion in Biotechnology, 14, 395–400. (a) van de Velde, F., Arends, I.W. and Sheldon, R.A. (2000) Journal of Inorganic Biochemistry, 80, 81–89. (b) Wahler, D. and Reymond, J.-L. (2002) Canadian Journal of Chemistry, 80, 665–670. (a) Thomas, C.M. and Ward, T.R. (2005) Chemical Society Reviews, 34, 337–346. (b) Letondor, C. and Ward, T.R. (2006) Chembiochem: A European Journal of Chemical Biology, 7, 1845–1852. (a) Bradley, L.H., Thumfort, P.P. and Hecht, M.H. (2006) Methods in Molecular Biology, 340, 53–69. (b) Berkessel, A. (2003) Current Opinion in Chemical Biology, 7, 409–419. Kofoed, J. and Reymond, J.-L. (2005) Current Opinion in Chemical Biology, 9, 656–664. (a) Stewart, J.D. and Benkovic, S.J. (1995) Nature, 375, 388–391. (b) Reymond, J.-L. and Chen, Y. (1996) Israel Journal of Chemistry, 36, 199. Tanaka, F. and Barbas, C.F. (2002) Journal of Immunological Methods, 269, 67–79. Reymond, J.-L., Janda, K.D. and Lerner, R.A. (1992) Journal of the American Chemical Society, 114, 2257. Reymond, J.-L., Jahanghiri, G.K. and Lerner, R.A. (1993) Journal of the American Chemical Society, 115, 3909. Reymond, J.-L., Reber, J.-L. and Lerner, R.A. (1994) Angewandte Chemie International Edition in English, 33, 475. Zheng, L., Baumann, U. and Reymond, J.-L. (2004) Proceedings of the National

20

21

22

23

24

25

26

27

28

29

30

Academy of Sciences of the United States of America, 101, 3387–3392. Zheng, L., Baumann, U. and Reymond, J.-L. (2003) Journal of Applied Biochemistry, 5, 577–581. (a) Wagner, J., Lerner, R.A. and Barbas, C.F. III (1995) Science, 270, 1797. (b) Zhong, G., Lerner, R.A. and Barbas, C.F. III (1999) Angewandte Chemie International Edition, 38, 3738. (c) Tanaka, F., Lerner, R.A. and Barbas, C.F. III (2000) Journal of the American Chemical Society, 122, 4835. Sinha, S.C., Sun, J., Miller, G.P., Wartmann, M. and Lerner, R.A. (2001) Chemistry, 7, 1691–1702. Khersonsky, O., Roodveldt, C. and Tawﬁk, D.S. (2006) Current Opinion in Chemical Biology, 10, 498–508. Svedendahl, M., Hult, K. and Berglund, P. (2005) Journal of the American Chemical Society, 127, 17988–17999. Hollfelder, F., Kirby, A.J., Tawﬁk, D.S., Kikuchi, K. and Hilvert, D. (2000) Journal of the American Chemical Society, 122, 1022–1029. Reetz, M.T. and Jiao, N. (2006) Angewandte Chemie International Edition, 45, 2416–2419. Roelfes, G. and Feringa, B.L. (2005) Angewandte Chemie International Edition, 44, 3230. Wilson, M.E. and Whitesides, G.M. (1978) Journal of the American Chemical Society, 100, 306. (a) Collot, J., Gradinaru, J., Humbert, N., Skander, M., Zocchi, A. and Ward, T.R. (2003) Journal of the American Chemical Society, 125, 9030. (b) Klein, G., Humbert, N., Gradinaru, J., Ivanova, A., Gilardoni, F., Rusbandi, U.E. and Ward, T.R. (2005) Angewandte Chemie International Edition, 44, 7764. (a) Kirby, A.J. (1996) Angewandte Chemie International Edition, 35, 706–724. (b) Woggon, W.-D. (2005) Accounts of Chemical Research, 38, 127–136.

j81

j 3 The Search for New Enzymes

82

31 (a) Newkome, G.R., Mooreﬁeld, C.N. and V€ ogtle, F. (2001) Dendrimers and Dendrons. Concepts. Synthesis and Perspectives, WileyVCH, New York. (b) Tomalia, D.A. and Dvornic, P.R. (1994) Nature, 372, 617–618. (c) Liang, C.O. and Frechet, J.M.J. (2005) Progress in Polymer Science, 30, 385–402. 32 Darbre, T. and Reymond, J.-L. (2006) Accounts of Chemical Research, 39, 925–934. 33 (a) Clouet, A., Darbre, T. and Reymond, J.-L. (2004) Angewandte Chemie International Edition, 43, 4612–4615. (b) Clouet, A., Darbre, T. and Reymond, J.L. (2006) Biopolymers, 84, 114–123. 34 (a) Fonseca, M.H. and List, B. (2004) Current Opinion in Chemical Biology, 8, 319–326. (b) Stambuli, J.P. and Hartwig, J.F. (2003) Current Opinion in Chemical Biology, 7, 420–426. 35 (a) Douat-Casassus, C., Darbre, T. and Reymond, J.-L. (2004) Journal of the American Chemical Society, 126, 7817–7826. (b) Delort, E., Darbre, T. and Reymond, J.-L. (2004) Journal of the American Chemical Society, 126, 15642–15643. (c) Delort, E., NguyenTrung, N.-Q., Darbre, T. and Reymond, J.-L. (2006) Journal of Organic Chemistry, 71, 4468–4480. 36 Kofoed, J., Darbre, T. and Reymond, J.-L. (2006) Organic and Biomolecular Chemistry, 3268–3281. 37 (a) Barbas, C.F., Wang, Y.F. and Wong, C.H. (1990) Journal of the American Chemical Society, 112, 2013–2014. (b) Heine, A., Luz, J.G., Wong, C.H. and Wilson, I.A. (2004) Journal of Molecular Biology, 343, 1019–1034. 38 (a) Kofoed, J., Nielsen, J. and Reymond, J.-L. (2003) Bioorganic and Medicinal Chemistry Letters, 13, 2445–2447. (b) Martin, H.J. and List, B. (2003) Synlett, 1901–1902. (c) Krattiger, P., Kovasy, R., Revell, J.D., Ivan, S. and Wennemers, H. (2005) Organic Letters, 7, 1101–1103.

39 Schmidt, T.M., DeLong, E.F. and Pace, N.R. (1991) Journal of Bacteriology, 173, 4371–4378. 40 Tsai, Y.L. and Olson, B.H. (1992) Applied and Environmental Microbiology, 58, 2292–2295. 41 (a) Amann, R.I., Ludwig, W. and Schleifer, K.H. (1995) Microbiological Reviews, 59, 143–169. (b) Handelsman, J., Rondon, M.R., Brady, S.F., Clardy, J. and Goodman, R.M. (1998) Chemistry and Biology, 5, R245–R249 (c) Hugenholtz, P., Goebel, B.M. and Pace, N.R. (1998) Journal of Bacteriology, 180, 4765–4774. 42 (a) Curtis, T.P. and Sloan, W.T. (2005) Science, 309, 1331–1333. (b) Daniel, R. (2005) Nature Reviews Microbiology, 3, 470–478. (c) Gans, J., Wolinsky, M. and Dunbar, J. (2005) Science, 309, 1387–1390. 43 (a) Lorenz, P. and Eck, J. (2005) Nature Reviews Microbiology, 3, 510–516. (b) Steele, H.L. and Streit, W.R. FEMS Microbiology Letters, 247, 105–111. 44 Brennan, Y., Callen, W.N., Christoffersen, L., Dupree, P., Goubet, F., Healey, S., Hernandez, M., Keller, M., Li, K., Palackal, N., Sittenfeld, A., Tamayo, G., Wells, S., Hazlewood, G.P., Mathur, E.J., Short, J.M., Robertson, D.E. and Steer, B.A. (2004) Applied and Environmental Microbiology, 70, 3609–3617. 45 Ferrer, M., Golyshina, O.V., Chernikova, T.N., Khachane, A.N., Martins dos Santos, V.A.P., Yakimov, M.M., Timmis, K.N. and Golyshin, P.N. (2005) Chemistry and Biology, 12, 895–904. 46 Courtois, S., Cappellano, C.M., Ball, M., Francou, F.-X., Normand, P., Helynck, G., Martinez, A., Kolvek, S.J., Hopke, J., Osburne, M.S., August, P.R., Nalin, R., Guerineau, M., Jeannin, P., Simonet, P. and Pernodet, J.-L. (2003) Applied and Environmental Microbiology, 69, 49–55. 47 Entcheva, P., Liebl, W., Johann, A., Hartsch, T. and Streit, W.R. (2001) Applied and Environmental Microbiology, 67, 89–99. 48 Rondon, M.R., August, P.R., Bettermann, A.D., Brady, S.F., Grossman, T.H., Liles,

References

49

50

51

52

53

54

55

56

M.R., Loiacono, K.A., Lynch, B.A., MacNeil, I.A., Minor, C., Tiong, C.L., Gilman, M., Osburne, M.S., Clardy, J., Handelsman, J. and Goodman, R.M. (2000) Applied and Environmental Microbiology, 66, 2541–2547. Kim, Y.-J., Choi, G.-S., Kim, S.-B., Yoon, G.-S., Kim, Y.-S. and Ryu, Y.-W. (2006) Protein Expression and Puriﬁcation, 45, 315–323. Henne, A., Daniel, R., Schmitz, R.A. and Gottschalk, G. (1999) Applied and Environmental Microbiology, 65, 3901–3907. Ranjan, R., Grover, A., Kapardar, R.K. and Sharma, R. (2005) Biochemical and Biophysical Research Communications, 335, 57–65. Gabor, E.M., de Vries, E.J. and Janssen, D.B. (2004) Environmental Microbiology, 6, 948–958. Venter, J.C., Remington, K., Heidelberg, J.F., Halpern, A.L., Rusch, D., Eisen, J.A., Wu, D., Paulsen, I., Nelson, K.A., Nelson, W., Fouts, D.E., Levy, S., Knap, A.H., Lomas, M.W., Nealson, K., White, O., Peterson, J., Hoffman, J., Parsons, R., Baden-Tillson, H., Pfannkoch, C., Rogers, Y.-H. and Smith, H.O. (2004) Science, 304, 66–74. Leggewie, C., Henning, H., Schmeisser, C., Streit, W.R. and Jaeger, K.-E. (2006) Journal of Biotechnology, 123, 281–287. (a) Li, Y., Wexler, M., Richardson, D.J., Bond, P.L. and Johnston, A.W.B. (2005) Environmental Microbiology, 7, 1927–1936. (b) Martinez, A., Kolvek, S.J., Yip, C.L.T., Hopke, J., Brown, K.A., MacNeil, I.A. and Osburne, M.S. (2004) Applied and Environmental Microbiology, 70, 2452–2463. (c) Wexler, M., Bond, P.L., Richardson, D.J. and Johnston, A.W. (2005) Environmental Microbiology, 7, 1917–1926. Wang, G.-Y.-S., Graziani, E., Waters, B., Pan, W., Li, X., McDermott, J., Meurer, G., Saxena, G., Andersen, R.J. and Davies, J. (2000) Organic Letters, 2, 2401–2404.

57 Uchiyama, T., Abe, T., Ikemura, T. and Watanabe, K. (2005) Nature Biotechnology, 23, 88–93. 58 Voget, S., Leggewie, C., Uesbeck, A., Raasch, C., Jaeger, K.-E. and Streit, W.R. (2003) Applied and Environmental Microbiology, 69, 6235–6242. 59 Ferrer, M., Golyshina, O.V., Chernikova, T.N., Khachane, A.N., Reyes-Duarte, D., Martins dos Santos, V.A.P., Strompl, C., Elborough, K., Jarvis, G., Neef, A., Yakimov, M.M., Timmis, K.N. and Golyshin, P.N. (2005) Environmental Microbiology, 7, 1996–2010. 60 Richardson, T.H., Tan, X., Frey, G., Callen, W., Cabell, M., Lam, D., Macomber, J., Short, J.M., Robertson, D.E. and Miller, C. (2002) Journal of Biological Chemistry, 277, 26501–26507. 61 Yun, J., Kang, S., Park, S., Yoon, H., Kim, M.-J., Heu, S. and Ryu, S. (2004) Applied and Environmental Microbiology, 70, 7229–9235. 62 Rothschild, L.J. and Mancinelli, R.L. (2001) Nature, 409, 1092–1101. 63 Grant, S., Sorokin, D.Y., Grant, W.D., Jones, B.E. and Heaphy, S. (2004) Extremophiles, 8, 421–429. 64 Healy, F.G., Ray, R.M., Aldrich, H.C., Wilkie, A.C., Ingram, L.O. and Shanmugam, K.T. (1995) Applied Microbiology and Biotechnology, 43, 667–674. 65 Voget, S., Steele, H.L. and Streit, W.R. (2006) Journal of Biotechnology, 126, 26–36. 66 Howard, M.B., Ekborg, N.A., Weiner, R.M. and Hutcheson, S.W. (2003) Journal of Industrial Microbiology and Biotechnology, 30, 627–635. 67 Carstens, M., Vivier, M.A., van Rensburg, P. and Pretorius, I.S. (2003) Annals of Microbiology, 53, 15–28. 68 (a) Cottrell, M.T., Moore, J.A. and Kirchman, D.L. (1999) Applied and Environmental Microbiology, 65, 2553–2557. (b) Cottrell, M.T., Wood, D.N., Yu, L. and Kirchman, D.L. (2000) Applied and Environmental Microbiology, 66, 1195–1201.

j83

j 3 The Search for New Enzymes

84

69 Lee, C.C., Kibblewhite-Accinelli, R.E., Wagschal, K., Robertson, G.H. and Wong, D.W.S. (2006) Extremophiles, 10, 295–300. 70 Henne, A., Schmitz, R.A., Bomeke, M., Gottschalk, G. and Daniel, R. (2000) Applied and Environmental Microbiology, 66, 3113–3116. 71 Rees, H.C., Grant, S., Jones, B., Grant, W.D. and Heaphy, S. (2003) Extremophiles, 7, 415–421. 72 Lee, S.W., Won, K., Lim, H.K., Kim, J.C., Choi, G.J. and Cho, K.Y. (2004) Applied Microbiology and Biotechnology, 65, 720–726. 73 Rhee, J.-K., Ahn, D.-G., Kim, Y.-G. and Oh, J.-W. (2005) Applied and Environmental Microbiology, 71, 817–825. 74 Kim, J.N., Seo, M.J., Cho, E.A., Lee, S.J., Kim, S.B., Cheigh, C.I. and Pyun, Y.R. (2005) Journal of Microbiology and Biotechnology, 15, 1067–1072. 75 Elend, C., Schmeisser, C., Leggewie, C., Babiak, P., Carballeira, J.D., Steele, H.L., Reymond, J.L., Jaeger, K.E. and Streit, W.R. (2006) Applied and Environmental Microbiology, 72, 3637–3645. 76 Eschenfeldt, W.H., Stols, L., Rosenbaum, H., Khambatta, Z.S., Quaite-Randall, E., Wu, S., Kilgore, D.C., Trent, J.D. and Donnelly, M.I. (2001) Applied and Environmental Microbiology, 67, 4206–4214. 77 Cantarella, M., Cantarella, L., Gallifuoco, A. and Spera, A. (2006) Enzyme and Microbial Technology, 38, 126–134. 78 Brady, D., Beeton, A., Zeevaart, J., Kgaje, C., van Rantwijk, F. and Sheldon, R.A. (2004) Applied Microbiology and Biotechnology, 64, 76–85. 79 Liebeton, K. and Eck, J. (2004) Engineering in Life Sciences, 4, 557–562. 80 Gabor, E.M., Alkema, W.B.L. and Janssen, D.B. (2004) Environmental Microbiology, 6, 879–886. 81 (a) DeSantis, G., Zhu, Z., Greenberg, W.A., Wong, K., Chaplin, J., Hanson, S.R., Farwell, B., Nicholson, L.W., Rand, C.L., Weiner, D.P., Robertson, D.E. and Burk,

82

83

84

85

86

87

M.J. (2002) Journal of the American Chemical Society, 124, 9024–9025. (b) Podar, M., Eads, J.R. and Richardson, T.H. (2005) BMC Evolutionary Biology, 5, 42. (c) Robertson, D.E., Chaplin, J.A., DeSantis, G., Podar, M., Madden, M., Chi, E., Richardson, T., Milan, A., Miller, M., Weiner, D.P., Wong, K., McQuaid, J., Farwell, B., Preston, L.A., Tan, X., Snead, M.A., Keller, M., Mathur, E., Kretz, P.L., Burk, M.J. and Short, J.M. (2004) Applied and Environmental Microbiology, 70, 2429–2436. Wang, C., Meek, D.J., Panchal, P., Boruvka, N., Archibald, F.S., Driscoll, B.T. and Charles, T.C. (2006) Applied and Environmental Microbiology, 72, 384–391. (a) Knietsch, A., Waschkowitz, T., Bowien, S., Henne, A. and Daniel, R. (2003) Applied and Environmental Microbiology, 69, 1408–1416. (b) Knietsch, A., Waschkowitz, T., Bowien, S., Henne, A. and Daniel, R. (2003) Journal of Molecular Microbiology and Biotechnology, 5, 46–56. Gupta, R., Beg, Q.K. and Lorenz, P. (2002) Applied Microbiology and Biotechnology, 59, 15–22. Knietsch, A., Bowein, S., Whited, G., Gottschalk, G. and Daniel, R. (2003) Applied and Environmental Microbiology, 69, 3048–3060. (a) Lim, H.K., Chung, E.J., Kim, J.-C., Choi, G.J., Jang, K.S., Chung, Y.R., Cho, K.Y. and Lee, S.-W. (2005) Applied and Environmental Microbiology, 71, 7768–7777. (b) MacNeil, I.A., Tiong, C.L., Minor, C., August, P.R., Grossman, T.H., Loiacono, K.A., Lynch, B.A., Phillips, T., Narula, S., Sundaramoorthi, R., Tyler, A., Aldredge, T., Long, H., Gilman, M., Holt, D. and Osburne, M.S. (2001) Journal of Molecular Microbiology and Biotechnology, 3, 301–308. (a) Brady, S.F. and Clardy, J. (2004) Journal of Natural Products, 67, 1283–1286. (b) Brady, S.F., Chao, C.J. and Clardy, J. (2004) Applied and Environmental Microbiology, 70, 6865–6870. (c) Brady, S.F. and Clardy, J. (2005) Angewandte Chemie

References International Edition, 44, 7063–7065. (d) Gillespie, D.E., Brady, S.F., Bettermann, A.D., Cianciotto, N.P., Liles, M.R., Rondon, M.R., Clardy, J., Goodman, R.M. and Handelsman, J. (2002) Applied and Environmental Microbiology, 68, 4301–4306.

88 (a) Diaz-Torres, M.L., McNab, R., Spratt, D.A., Villedieu, A., Hunt, N., Wilson, M. and Mullany, P. (2003) Antimicrobial Agents and Chemotherapy, 47, 1430–1432. (b) Riesenfeld, C.S., Goodman, R.M. and Handelsman, J. (2004) Environmental Microbiology, 6, 981–989.

j85

j87

II Synthetic Applications

Asymmetric Organic Synthesis with Enzymes. Edited by Vicente Gotor, Ignacio Alfonso, and Eduardo Garcıa-Urdiales Copyright Ó 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 978-3-527-31825-4

j89

4 Dynamic Kinetic Resolutions Belen Martın-Matute and Jan-E. B€ackvall

4.1 Introduction 4.1.1 Synthesis of Enantiomerically Pure Compounds

There are three strategies for preparing enantiomerically pure compounds [1]. One way is to use naturally occurring starting materials of deﬁned absolute conﬁguration, provided by natures chiral pool. The second approach is to perform a resolution of a racemate. Thirdly, an asymmetric synthesis can be achieved in which one or more chiral centers are created in an achiral starting material. The ﬁrst two approaches have some important drawbacks: in the ﬁrst approach, an adequate starting material must be available that can be efﬁciently transformed to the desired product in a few steps, and in the second approach, the maximum theoretical yield is only 50%. For these reasons, asymmetric synthesis accomplished by the use of a chiral catalyst has been considered as the most reﬁned strategy. In this approach, one enantiopure molecule (catalyst) is able to transform many prochiral molecules in enantiopure products (chiral multiplication). The chiral ligand can be an enzyme, a simple organic molecule, or a metal complex bearing chiral ligands. There are some examples of metal-catalyzed asymmetric transformations that have been developed to produce chiral compounds on a big scale. For example, in the Monsanto process developed by Knowles, chiral diphosphines are employed as chiral ligands in an Rh-catalyzed hydrogenation [2]. Another example is the asymmetric hydrogenation developed by Noyori in which Ru complexes bearing chiral phosphines are used [3]. Despite the still growing number of available methods for the preparation of enantiopure compounds by the use of asymmetric catalysis, kinetic resolution (KR) is still the most employed method in the industry [4], and in most cases biocatalysts (enzymes) are used.

Asymmetric Organic Synthesis with Enzymes. Edited by Vicente Gotor, Ignacio Alfonso, and Eduardo Garcıa-Urdiales Copyright Ó 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 978-3-527-31825-4

j 4 Dynamic Kinetic Resolutions

90

cat* s Fast

p 50%

ent- s

cat* ent- p Slow

Figure 4.1 Kinetic resolution (s is substrate, p is product, cat* is homochiral catalyst).

4.1.2 Kinetic Resolution (KR) and Dynamic Kinetic Resolution (DKR)

KR is the total or partial separation of two enantiomers from a racemic mixture [5]. KR is based on the different reaction rates of the enantiomers with a chiral molecule (a reagent, a catalyst, etc). In the ideal case, the difference in reactivity is large, and one of the enantiomers reacts very fast to give the product, whereas the other does not react at all (Figure 4.1). Many enzyme-catalyzed KRs do not show this ideal behavior. Usually the reaction does not stop at 50% conversion, but it only slows down. The concentration of each enantiomer is not constant during the reaction, and therefore, the reaction rate of both enantiomers changes with the conversion. To obtain the product with high enantiomeric excess, it is necessary to stop the reaction before it reaches 50% conversion. In a KR the enantiomeric excess of the substrate and product change with the conversion [6]. A direct comparison of the enantiomeric excess of two KRs is relevant only if they have the same conversion. Owing to the intrinsic difﬁculty that this implies, Sih developed a series of equations to calculate the enantioselectivity (E ) of an enzyme-catalyzed KR [7]. The enantioselectivity known as enantiomeric ratio (E ¼ ks/kents) (for biocatalyzed reactions, the binding of the substrate enantiomers (which can be neglected with chemical catalysts) usually plays an important role in the chiral selection process and E values of enzyme-catalyzed reactions are therefore deﬁned through Michaelis–Menten kinetics: E ¼ (kcat/KM)R/ (kcat/KM)S), measures the ability of an enzyme to distinguish between two enantiomers [8]. E remains constant throughout the reaction, and allows easy comparison between two KRs. It can be calculated very easily knowing two of the three following parameters: conversion (c), enantiomeric excess of the product (eep), and enantiomeric excess of the substrate (ees) (Faber has developed a simple program (Selectivity) that allows to calculate the E value and to draw plots of the variation of ees and eep as a function of time for an irreversible kinetic resolution. The program can be obtained on: http://borgc185.kfunigraz.ac.at/). As a rule of thumb KRs with E values smaller than 15 do not have applications in synthesis. Values of E of 15–30 are considered as moderate or good, and values >30 are regarded as excellent with useful applications in organic synthesis [1f ]. Enzymatic KRs, as all resolutions, are limited to a maximum theoretical yield of 50%. Strategies to increase the yield are therefore of great importance. The opposite of a resolution, that is, the racemization of a chiral compound, can sometimes be highly desirable and applicable in enantioselective synthesis. By combining a

4.1 Introduction

s Racemization krac ent-s

cat* ks Fast cat*

p 100 %

ent-p

kent-s Slow Figure 4.2 Dynamic kinetic resolution.

racemization with an enzyme-catalyzed resolution, a highly efﬁcient asymmetric transformation to only one enantiomer can be obtained. Such dynamic kinetic resolutions (DKRs) with a theoretical yield of 100% are a powerful approach to prepare enantiomerically pure molecules (Figure 4.2). Racemization can be performed by a chemocatalyst or a biocatalyst, or it can occur spontaneously. For an efﬁcient enzymatic DKR the following requirements must be fulﬁlled: (i) the KR must be very selective (E > 20); (ii) the racemization must be fast (at least 10 times faster than the enzyme-catalyzed transformation of the slow reacting enantiomer, krac >10 kents); (iii) the racemization catalyst must not react with the product of the reaction; (iv) the KR and the racemization must be compatible under the same reaction conditions. In an ideal DKR, where the substrate stays racemic throughout the reaction process, the optical purity depends only on the enantiomeric ratio (E ) (ee ¼ (E 1)/ (E þ 1)), and is independent of the extent of conversion. The enantiomeric excess of the product formed under racemizing conditions is equal to the initial enantiomeric excess under nonracemizing conditions. 4.1.3 Enzymes in Organic Chemistry

Biocatalysts usually require mild reaction conditions for an optimal activity (physiologic temperature and pH) and, in general, they show high activity, chemo- and enantioselectivity. Furthermore, when using enzymes, many functional group protections and/or activations can be avoided, allowing shorter synthetic transformations. The use of enzymes is therefore very attractive from an environmental and economic point of view. Lipases and proteases belong to the group of hydrolases [1e,9]. Hydrolases are excellent catalysts and many of them are commercially available. They do not require a cofactor and they catalyze the hydrolysis of nonnatural esters very efﬁciently. However, their use in organic synthesis dates from only two decades ago. Since then, the number of reports on enzyme-catalyzed synthetic transformations has grown exponentially. Before that period, hydrolases were only employed in aqueous media, which is a drawback in synthesis owing to the insolubility of organic compounds in water. During the 1980s it was discovered that enzymes can be employed in organic media since they keep their native structures even in

j91

j 4 Dynamic Kinetic Resolutions

92

anhydrous organic solvents [10]. This stimulated the development of many new enzyme-catalyzed transformations such as transesteriﬁcations, aminolysis, or thiotransesteriﬁcations. One of the most important developments, the combination of enzymatic resolution with racemization (to give a DKR), could also be realized because of the possibility of running enzymatic reactions in organic solvents. In this chapter, DKRs will be categorized according to the racemization method employed, as being catalyzed by (i) a metal, (ii) a base, (iii) an acid, (iv) an aldehyde, or (v) an enzyme. Also racemizations that take place through continuous cleavage/ formation of the substrate, or through SN2 displacement, among other methods, will be discussed. In most cases, the racemization method of choice depends on the structure of the substrate. In all cases, the KR is catalyzed by an enzyme.

4.2 Metal-Catalyzed Racemization

A range of metal-catalyzed racemization reactions are known in the literature, but it is not always straightforward to combine these with an enzymatic resolution. For example, several complexes of rhodium, iridium, and ruthenium, among others, are known to catalyze rapid racemization of a variety of alcohol substrates [11], but only few have proved compatible with an enzymatic resolution. Most often, the compatibility of the two catalysts under the same reaction conditions is the major problem: the metal may interfere with the enzyme to give poor resolution or the enzyme may slow down or inhibit the racemization by the metal catalyst. Furthermore, the byproducts produced in the metal- and/or enzyme-catalyzed reaction or the surfactants employed to support the enzyme can also interfere with the racemization or with the enzymatic resolution. Another problem that has to be overcome is to ﬁnd the appropriate solvent for optimal activity of both catalysts, the enzyme and the metal complex: lipases show excellent activity and selectivity in water, but most organometallic complexes do not work in water. Lipases also work very well in apolar organic solvents, such as hexane. However, most organometallic complexes are not soluble in very apolar solvents, resulting in a low rate of the racemization reaction. Also, racemization is usually faster at high temperatures, at which most enzymes lose their selectivity and catalytic activity. Despite this, a deep understanding concerning the compatibility of the two catalysts has been achieved, and excellent results have been obtained. In 1997, St€ urmer already highlighted the importance of the combination of enzymes and transition metals in one pot [12]. Since then, this concept has attracted considerable interest within the scientiﬁc community. In all the DKRs presented in this section, the enzyme catalyzes a transesteriﬁcation process. Thus, enzyme- and metal-catalyzed DKRs will be categorized according to the nature of the substrate as being allylic substrates, secondary alcohols, or primary amines. In the ﬁrst case, racemization occurs through 1,3-shift of an acetate or a hydroxyl group, and in the last two cases it occurs through hydrogen transfer processes.

4.2 Metal-Catalyzed Racemization

OAc

OAc Ph

PrOAc

(R)

THF, 25 C, 1.5 d

Ph

71% yield, 98% ee

Ph

Pd AcO

i

o

–Pd(0)

+Pd(0)

OH

CALB, iPrOH

Ph

j93

L

Figure 4.3 DKR of allylic acetates catalyzed by a lipase and Pd(0).

4.2.1 DKR of Allylic Acetates and Allylic Alcohols

The ﬁrst example of chemoenzymatic DKR of allylic alcohol derivatives was reported by Allen and Williams [13]. Cyclic allylic acetates were deracemized by combining a lipasecatalyzed hydrolysis with a racemization via transposition of the acetate group catalyzed by a Pd(II) complex. Despite the limitation of the process with very long reaction times (19 days), this work was a signiﬁcant step forward in the combination of enzymes and metals in one pot. Some years later, Kim et al. considerably improved the DKR of allylic acetates using a Pd(0) complex for the racemization, which occurs through p-(allylpalladium) intermediates. The transesteriﬁcation is catalyzed by a lipase (Candida antarctica lipase B (CALB)) using isopropanol as acyl acceptor (Figure 4.3) [14]. A novel approach was developed very recently by Kita et al. [15]. DKR of allylic alcohols was performed by combining a lipase-catalyzed acylation with a racemization through the formation of allyl vanadate intermediates. Excellent yields and enantioselectivities were obtained. An example is shown in Figure 4.4. A limitation with this approach for the substrates shown in Figure 4.4 is that the allylic alcohol must be equally disubstituted in the allylic position (R1 ¼ R2) since CC single bond rotation is required in the tertiary alkoxy intermediate. Alternatively, R1 or R2 can be H if the two allylic alcohols formed by migration of the hydroxyl group are enantiomers (e.g. cyclic allylic acetates). DKR of allylic alcohols can be also performed using ruthenium complexes for the racemization that occurs through hydrogen transfer reactions (vide infra) [16]. EtO OH

OAc

CALB,

OH

VO(OSiPh 3)3 (10 mol %) +[VO(OR)3]

–[VO(OR)3]

OAc (R)

Acetone, 25 oC, 4.5 d 91% yield, 99% ee

OR RO V O O R2

OR V O O

RO

R1

R2 1

R

OR RO V O O R2 R1

Figure 4.4 DKR of allylic alcohols via vanadate intermediates.

j 4 Dynamic Kinetic Resolutions

94

[M–H] + H

O

O

H R

R'

or

R

M

R'

H

O

H

O R

M

R

H R' R or S

H H

R' rac

Figure 4.5 Simplified mechanism of the racemization of secalcohols catalyzed by transition metal complexes.

4.2.2 DKR of sec-alcohols

The racemization mechanism of sec-alcohols has been widely studied [16,17]. Metal complexes of the main groups of the periodic table react through a direct transfer of hydrogen (concerted process), such as aluminum complexes in Meerwein–Ponndorf–Verley–Oppenauer reaction. However, racemization catalyzed by transition metal complexes occurs via hydrogen transfer processes through metal hydrides or metal dihydrides intermediates (Figure 4.5) [18]. The research groups of Williams [19] and B€ackvall [20] were the ﬁrst to report the combination of enzymes and transition metals for DKR of sec-alcohols. Williams employed complexes of Al, Rh, or Ir in combination with Pseudomonas ﬂuorescens lipase (PFL) for the DKR of 1-phenylethanol. The best results were obtained using Rh2(OAc)4 as the catalyst for the racemization, and 60% conversion of the alcohol to give 1-phenylethyl acetate in 98% ee was obtained (Figure 4.6) [19]. At higher conversion, the enantiomeric excess dropped and 76% conversion gave 80% ee. The research group of B€ackvall employed the Shvos ruthenium complex (1) [21] for the racemization. This complex is activated by heat. For the KR they used p-chlorophenyl acetate as the acyl donor in combination with thermostable enzymes, such as CALB [20] (Figure 4.7). This was the ﬁrst practical chemoenzymatic DKR affording acetylated sec-alcohols in high yields and excellent enantioselectivities. In the best case 100% conversion (92% isolated yield) with 99% ee was obtained. This method was subsequently applied to a variety of different substrates and it is employed (with a different ruthenium complex) by the Dutch company DSM for the large-scale production of (R)-phenylethanol [22]. Kim and Park subsequently reported that ruthenium precatalyst (2) racemizes alcohols within 30 minutes at room temperature [23]. However, when combined OH Ph

OAc Rh2(OAc)4 (2 mol %)

OAc Ph (R)

PFL, o-phenantroline (6 mol %), PhCOMe (1 equiv) 60% conversion, 98% ee Cyclohexane, 20 oC, 72 h

Figure 4.6 DKR of sec-alcohols catalyzed by PFL and Rh complexes.

4.2 Metal-Catalyzed Racemization Ph

O

j95 Ph

O H

OH

OAc

CALB, p-Cl-C6H4-OAc (3 equiv)

R

R (R)

[Ru] (1) (2 mol %) o

78–92% yield, >99% ee

Toluene, 70 C

R = Alkyl, aryl

Ph Ph

Ph Ph

Ru OC CO

H

Ru CO CO

1

24–72 h

Figure 4.7 DKR of sec-alcohols using CALB and Shvos complex (1).

with an enzyme (lipase) in DKR at room temperature, very long reaction times (1.3–7 days) were required, in spite of the fact that the enzymatic KR takes only a few hours (Figure 4.8). Despite these compatibility problems, their results constitute an important improvement, since chemoenzymatic DKR could now be performed at ambient temperature to give high yields, which enables the use of nonthermostable enzymes. More recently, we accomplished a highly efﬁcient metal- and enzyme-catalyzed DKR of alcohols at room temperature (Figure 4.8) [16,24]. This is the fastest DKR of alcohols hitherto reported by the combination of transition metal and enzyme catalysts. Racemization was effected by a very potent hydrogen transfer catalyst (3). Catalysts (2) and (3) have been employed in the DKR of a broad range of secondary alcohols. We have studied the racemization and found that it proceeds via a new mechanism where the intermediate ketone formed in the process stays coordinated to ruthenium [16]. Recently, the dynamic kinetic asymmetric transformation (DYKAT) of diols was improved signiﬁcantly by using the highly efﬁcient catalyst (3) to achieve very fast epimerization [25,26]. In the DYKATof 2,5-hexanediol catalyzed by (3) [25], the faster epimerization and the low reaction temperature reduce the concentration of intermediates that can evolve by anomalous enzyme-catalyzed (S)-acylations, such as alcohols bearing a carbonyl group or an (R)-acetoxy group at the d-position that give rise to the unwanted formation of meso-diacetates. In the DYKAT of 2,4-pentanediol catalyzed by (3), the fast epimerization outruns unwanted acyl migrations that yield meso-diacetates. Thus, by using complex (3) (R,R)-2,5-hexanediol diacetate and (R,R)-2,4-pentanediol diacetate can now be obtained in quantitative yield with

OH

CALB,

Ph

OAc

OAc Ph

[Ru] (4 mol %), KOt Bu (5 mol %), Na 2CO3 o

Ph

[Ru] = 2, 31 h [Ru] = 3, 3 h

H N

Ph

Ph Ph Ru Cl OC CO

2

(R)

95–99% yield, >99% ee

Toluene, 23 C

Ph

Ph

Ph

Ph

Ph Ph

Ph Ru Cl OC CO

3

Figure 4.8 DKR of sec-alcohols catalyzed by a lipase and Ru complexes at ambient temperature.

j 4 Dynamic Kinetic Resolutions

96

excellent enantiomeric excess (>99%) and high (R,R)/meso ratio (94:6 and 97:3, respectively). Kim and Park et al. have reported a polymer-supported derivative (4), which, together with the enzyme, can be recycled and reused [27].

O O Ph Ph

O Ph

Ph Ru Cl OC CO

4 Sheldon et al. have combined a KR catalyzed by CALB with a racemization catalyzed by a Ru(II) complex in combination with TEMPO (2,2,6,6-tetramethylpiperidine 1-oxyl free radical) [28]. They proposed that racemization involved initial ruthenium-catalyzed oxidation of the alcohol to the corresponding ketone, with TEMPO acting as a stoichiometric oxidant. The ketone was then reduced to racemic alcohol by ruthenium hydrides, which were proposed to be formed under the reaction conditions. Under these conditions, they obtained 76% yield of enantiopure 1-phenylethanol acetate at 70 after 48 hours. Naturally occurring lipases are (R)-selective for alcohols according to Kazlauskas rule [29] ((R)- and (S)-selective are used for typical sec-alcohols where the large group has a higher priority in the sequential rule for determining the conﬁguration (according to the Cahn–Ingold–Prelog system). Thus DKR of alcohols employing lipases can only be used to transform a racemic alcohol into the (R)-acetate. Serine proteases, a subclass of hydrolases, are known to catalyze transesteriﬁcations similar to those catalyzed by lipases, but interestingly, they often show the reversed enantioselectivity to give the (S)-product. Proteases are less thermostable enzymes, and for this reason metal complexes that racemize secondary alcohols at ambient temperature are desirable in DKR of sec-alcohols with these enzymes. Ruthenium complexes (2) and (3) have been combined with subtilisin Carlsberg, affording a method for the synthesis of acetylated (S)-sec-alcohols [30]. With complex (2) the DKR reaction took 3–4 days [30a], whereas with catalyst (3) DKR was obtained with reaction time down to 18 hours [30b], in both cases at room temperature. Very recently the Meerwein–Ponndorf–Verley–Oppenauer (MPVO) reaction has been exploited for the racemization of alcohols using inexpensive aluminum-based catalysts. Combination of these complexes with a lipase (CALB) results in an efﬁcient DKR of sec-alcohols at ambient temperature. To increase the reactivity of the aluminum complexes, a bidentate ligand, such as binol, is required. Also, speciﬁc acyl donors need to be used for each substrate [31] (Figure 4.9). Kita et al. have made use of the structure of the acyl donor to develop a domino transformation, a DKR followed by an intramolecular Diels–Alder reaction [32]. They

4.2 Metal-Catalyzed Racemization

OH

OAc

CALB, AlMe3 (10 mol %),

Ph

j97

Ph

OAc

(R)

96% yield, 96% ee (1.2 equiv) Binol (10 mol %) Toluene, rt, 3 h

Figure 4.9 DKR of sec-alcohols using a lipase and inexpensive Al complexes.

O OH

EtO

O

CO2Et

CO2Et

O O

O

O

CO2Et

CALB, Et3N, MS

Cl Ru Cl

81% yield, 97% ee 2

(10 mol %) MeCN, 35 oC, 3 d

Figure 4.10 Tandem DKR-intramolecular Diels–Alder reaction.

have found reaction conditions for three different transformations to effectively take place in one pot: a KR, a racemization, and a Diels–Alder reaction (Figure 4.10). Meijer and Heise and coworkers have developed iterative tandem catalysis with the aim of generating chiral oligomers [33]. The polymerization of 6-methyl-e-caprolactone was achieved by combining enzymatic ring opening with ruthenium-catalyzed racemization in one pot yielding enantioenriched oligomers (Figure 4.11). More recently, Heise and coworkers have shown that DKR can be combined with enzymatic polymerization for the synthesis of chiral polyesters from racemic secondary diols in one pot [34] (Figure 4.12). O OH

O

CALB

O O

O O

OH O n

CALB

O

Ru HN O

n = n +1

O

O O

OH O n

Figure 4.11 Synthesis of chiral oligomers by DKR.

NH

j 4 Dynamic Kinetic Resolutions

98

HO

O

OH

OMe

MeO O

Ru HN

CALB

NH O

O O O O

n

Figure 4.12 Synthesis of chiral polyesters by DYKAT.

4.2.3 DKR of Amines

Racemization of amines is difﬁcult to achieve and usually requires harsh reaction conditions. Reetz et al. developed the ﬁrst example of DKR of amines using palladium on carbon for the racemization and CALB for the enzymatic resolution [35]. This combination required long reaction times (8 days) to obtain 64% yield in the DKR of 1-phenylethylamine. More recently, B€ackvall et al. synthesized a novel Shvo-type ruthenium complex (5) that in combination with CALB made it possible to perform DKR of a variety of primary amines with excellent yields and enantioselectivities (Figure 4.13) [36]. Jacobs et al. have found that the efﬁciency of the Pd-catalyzed racemization of amines can be improved by using Pd immobilized on supports such as BaSO4, CaCO3, or BaCO3. The racemization was combined with a KR catalyzed by CALB affording enantiopure acetylated benzylamines in high yields [37]. Very recently Page and coworkers have reported the DKR of sec-amines using a low catalyst loading of an Ir complex for the racemization, and Candida rugosa lipase for the enzymatic resolution [38].

4.3 Base-Catalyzed Racemization

In this chapter, DKR of substrates having a proton with low pKa will be discussed. Racemization occurs by performing the DKR in the presence of a weak base. Also in

4.3 Base-Catalyzed Racemization

NH 2

OAc

CALB,

R

j99

NHAc R

[Ru] (5) (4 mol %), Na 2CO3 Toluene, 90 oC 3d

R = Ph, p-MeOC 6H 4

R

O

R = Ph, 90% yield, 98% ee R = p-MeOC6H4 , 95% yield, 99% ee R

O H

R

R R

Ru

R

R R

H

Ru OC CO CO CO 5 (R = p-MeO-C6 H4 ) Figure 4.13 Ru- and lipase-catalyzed DKR of primary amines.

this section, enzyme- and base-catayzed DKRs will be categorized according to the nature of the substrates as being thioesters, a-activated esters, oxazolones, hydantoins, or acyloins. 4.3.1 DKR of Thioesters

In contrast to oxoesters, the a-protons of thioesters are sufﬁciently acidic to permit continuous racemization of the substrate by base-catalyzed deprotonation at the acarbon. Drueckhammer et al. ﬁrst demonstrated the feasibility of this approach by performing DKR of a propionate thioester bearing a phenylthiogroup, which also contributes to the acidity of the a-proton (Figure 4.14) [39a]. The enzymatic hydrolysis of the thioester was coupled with a racemization catalyzed by trioctylamine. Owing to the insolubility of the substrate and base in water, they employed a biphasic system (toluene/H2O). Using P. cepacia (Amano PS-30) as the enzyme and a catalytic amount of trioctylamine, they obtained a quantitative yield of the corresponding

O PhS

O PhS

SEt

O

Pseudomonas cepacia

PhS

SEt

OH

Oct3N (0.5 equiv) Toluene / H2O, 25 oC 65 h

OPhS

SEt

Figure 4.14 DKR of thioesters using a base for racemization.

>99% yield, 96.3% ee

EtSH

j 4 Dynamic Kinetic Resolutions

100

Cl

Cl O

OH

CF3 Candida cylindr acea lipase O

Oct 3N, o Isooctane/H 2O, 30 C

O

+ CF3CH 2OH

94% yield, 99.5% ee

4d

Figure 4.15 DKR of activated esters using a base for racemization.

carboxylic acid in 96.3% ee. In this DKR, a maximum enantiomeric excess of 96.8% is expected, since the E value for the enzymatic resolution is 62.4. Drueckhammer et al. also reported DKR of thioesters not having activating groups at the a-position [39]. 4.3.2 DKR of Activated Esters

DKR of esters bearing an electron-withdrawing group at the a-carbon can be performed easily under mild reaction conditions due to the low pKa of the a-proton. Tsai et al. have reported an efﬁcient DKR of rac-2,2,2-triﬂuoroethyl a-chorophenyl acetate in water-saturated isooctane [40]. They used lipase MY from C. rugosa for the KR and trioctylamine as the base for racemization. (R)–chlorophenylacetic acid was obtained in 93% yield and 89.5% ee (Figure 4.15). 4.3.3 DKR of Oxazolones

A very attractive and efﬁcient method for the synthesis of L-aminoacids via DKR has been reported by Turner et al. [41a,b]. They employed enzyme-catalyzed ring opening of 5(4H)-oxazolones in combination with a catalytic amount of Et3N. The relatively low pKa of the C-4 proton (8.9) of oxazolones facilitates racemization. Hydrolysis of the ester obtained through DKR, followed by debenzoylation, yields L-aminoacids in excellent enantiomeric excess (99.5%) (Figure 4.16). In their initial studies, they employed Rhizomucor miehei lipase (Lipozyme) as the biocatalyst [41]. More recently, they have obtained excellent results employing CALB [41b]. This method has also been employed by Bevinakatti [41c,d] and Sih [41e,f ].

O N Ph

O

H HN Et3N (25 mol %), BuOH (2 equiv) o

Toluene, 30 C 5d

H O

Rhizomucor miehei lipase

Ph

OBu O

94% yield, 99.5% ee

Figure 4.16 DKR of oxazolones using a base for racemizing.

O H2N

OH

4.4 Acid-Catalyzed Racemization

NH2

O Ph

HN

HN

D-hydantoinase

NH

Ph Borate buffer o pH = 9, 40 C 2d

O

O CO2H

73% yield, >99% ee

Figure 4.17 DKR of hydantoins under weak basic conditions.

4.3.4 DKR of Hydantoins

Another approach for the synthesis of enantiopure amino acids or amino alcohols is the enantioselective enzyme-catalyzed hydrolysis of hydantoins. As discussed above, hydantoins are very easily racemized in weak alkaline solutions via keto enol tautomerism. Sugai et al. have reported the DKR of the hydantoin prepared from DL-phenylalanine. DKR took place smoothly by the use of D-hydantoinase at a pH of 9 employing a borate buffer (Figure 4.17) [42]. 4.3.5 DKR of Acyloins

Ogasawara et al. took advantage of the easy racemization of acyloins in the presence of a weak base for the DKR of endo-3-hydroxytricyco[4.2.1.02,5]non-7-en-4-one (Figure 4.18) [43]. Acylation of the hydroxyl group was catalyzed by a lipase, and racemization took place via a transient meso-enediol.

4.4 Acid-Catalyzed Racemization

In contrast to enzyme- and base-catalyzed DKRs, there are only a few reports of enzyme- and acid-catalyzed DKRs. A plausible explanation is that deactivation of the enzyme can occur under acidic conditions. Also, decomposition of the substrate has been observed.

O

OH H

+

H HO

O

OH H

Lipase, Et 3N Vinyl acetate

O 75% yield, 97% ee

OH OH Figure 4.18 DKR of acyloins through meso-enediol intermediates.

j101

j 4 Dynamic Kinetic Resolutions

102

O OH

OH

CALB,

O

H-beta zeolite CP814E-22

C7H15

Vinyl octanoate (16 equiv) Octane/H2O, 60 oC, 8 h

–H +, +H2O

+H+, –H2O

90% yield, >99% ee

Figure 4.19 DKR of sec-alcohols catalyzed by acid zeolites and a lipase.

Jacobs et al. employed an acidic zeolite catalyst for the racemization of sec-alcohols, which occurs through the formation of carbocations [44] (Figure 4.19). The KR is catalyzed by CALB in the presence of vinyl octanoate as acyl donor. DKR takes place successfully in a biphasic system (octane/H2O, 1 : 1) at 60 C. Another example of enzyme- and acid-catalyzed DKR has been reported by Bornscheuer [45]. Acyloins were racemized by using an acidic resin through the formation of enol intermediates. The enzymatic resolution was catalyzed by CALB. Since deactivation of this enzyme occurred in the presence of the acidic resin, they designed a simple reactor setup with two glass vials connected via a pump to achieve a spatial separation between the acidic resin and the enzyme (Figure 4.20).

4.5 Racemization through Continuous Reversible Formation–Cleavage of the Substrate

Racemization of some substrates can take place through reversible formation of the substrate via an addition/elimination process. The racemization can be acid or base catalyzed. In this section we will discuss DKR of cyanohydrins and hemithioacetals.

Two-compartment DKR

O

O

CALB, amberlist 15

Ph

O Ph

Ph

OH

OH

Vinyl butyrate Hexane, 40–50 oC

O O

>90% conversion, >91% ee

OH Ph OH Figure 4.20 DKR of sec-alcohols catalyzed by an acidic resin and a lipase.

4.5 Racemization through Continuous Reversible Formation–Cleavage of the Substrate

j103

4.5.1 DKR of Cyanohydrins

Several reports on DKR of cyanohydrins have been developed using this methodology. The unstable nature of cyanohydrins allows continuous racemization through reversible elimination/addition of HCN under basic conditions. The lipase-catalyzed KR in the presence of an acyl donor yields cyanohydrin acetates, which are not racemized under the reaction conditions. In 1992, Oda et al. reported a one-pot synthesis of optically active cyanohydrin acetates from aldehydes, which were converted to the corresponding racemic cyanohydrins through transhydrocyanation with acetone cyanohydrin, catalyzed by a a strongly basic anion-exchange resin [46]. The racemic cyanohydrins were acetylated by a lipase from P. cepacia (Amano) with isopropenyl acetate as the acyl donor. The reversible nature of the base-catalyzed transhydrocyanation enabled continuous racemization of the unreacted cyanohydrins, thereby effecting a total conversion (Figure 4.21). Kanerva et al. have also reported DKR of cyanohydrins [47]. In particular, they obtained very good results with C. antartica lipase A (CAL-A) as the catalyst for the KR of a variety of substrates for which other enzymes such as CALB or PS-C do not give good results (Figure 4.22) [47a]. Burk and coworkers have used a variety of nitrilases for the DKR of cyanohydrins [48]. Nitrilases catalyze the hydrolytic conversion of cyanohydrins directly to the corresponding carboxylic acids. Racemization was performed under basic conditions (phosphate buffer, pH 8) through reversible loss of HCN. (R)-Mandelic acid was obtained in high yield (86% yield) and high enantioselectivity (98% ee) after 3 hours (Figure 4.23).

4.5.2 DKR of Hemithioathetals

As in the case discussed above, hemithioacetals can be racemized by elimination/ addition of a small molecule (a thiol in this case) under weak acidic conditions. Rayner et al. reported the ﬁrst example of DKR of this kind of substrates yielding homochiral a-acetoxysulﬁdes (Figure 4.24) [49].

NC

O

OH

H

CN Anion-exchange resin

OAc

OH Pseudomonas cepacia lipase

OAc

(OH- form) (10 mol %) (3 equiv)

Figure 4.21 DKR of cyanohydrins.

CN

>96% yield, 84% ee 6.5 d

j 4 Dynamic Kinetic Resolutions

104

NC

OH

N

N

S

O

CHO

S

Amberilite IRA 904 (OH– form)(1 x 10 –5 mol %)

CN

O HO

CAL-A. Acetonitrile

AcO (2equiv)

N S

CN

O AcO

91% yield, 96% ee 48 h

Figure 4.22 Lipase-catalyzed DKR of cyanohydrins.

OH

OH

Nitrilase, MeOH

CN

CO2H

Phosphate buffer pH = 8, 37 oC 3h

86% yield, 98% ee

Figure 4.23 Nitrilase-catalyzed DKR of cyanohydrins.

O O O

H

OH

C8H17SH SiO2

O

SC8H17

O

OAc

Pseudomonas fluorescens lipase

AcO t-Butyl methyl ether, 30 oC

O

SC8H17

O 85% yield, >95% ee

6d

Figure 4.24 DKR of hemithioacetals.

4.6 Racemization Catalyzed by Aldehydes

In 1983, Yamada et al. developed an efﬁcient method for the racemization of amino acids using a catalytic amount of an aliphatic or an aromatic aldehyde [50]. This method has been used in the DKR of amino acids. Figure 4.25 shows the mechanism of the racemization of a carboxylic acid derivative catalyzed by pyridoxal. Racemization takes place through the formation of Schiff-base intermediates. In 1994, Wang et al. reported the DKR of amino acid derivatives by using pyridoxal 5-phosphate as the racemization catalyst [51]. The enzyme employed to catalyze the

4.6 Racemization Catalyzed by Aldehydes

N HOH2C NH2

H2O

OH N

N X

X

R O

O N N

CH3

HOH2C

OH

H2O

CH3 OH

O HOH2C

NH2 X

R

N

O

X

R

CH3

HOH2C

OH

R

O

N

CH3

HOH2C

OH O

X

R

N

CH3

O Figure 4.25 Racemization of amino acids through formation of Schiff-base intermediate.

KR was an alcalase subtilisin Carlsberg as the major enzyme component). To avoid racemization of the ﬁnal product, they employed a mixture of 2-methyl-2-propanol/ H2O (19 : 1). Under these conditions, the product precipitated during the course of hydrolysis (Figure 4.26). A very similar DKR process was reported two years later by Parmar [52]. Sheldon et al. have reported the DKR of phenylglycine esters via lipase-catalyzed ammonolysis [53]. Racemization was carried out by an aldehyde, such as salicyaldehyde or pyridoxal, under basic conditions. The major problem they found was the racemization by these aldehydes of the ﬁnal products. However, when performing the DKR at low temperatures (20 C) the substrate was racemized much faster than the product, and DKR was feasible yielding the product in good yield and high enantiomeric excess (Figure 4.27). A very elegant approach has been developed by Kanerva et al. DKR of N-hetrocyclic a-amino esters is achieved using CAL-A [54]. Racemization occurs when acetaldehyde is released in situ from the acyl donor. In this case aldehyde-catalyzed racemization of the product cannot occur (Figure 4.28). This is one of the few examples reported for DKR of secondary amines (For a recent example see the above text and Ref. [38]). O O NH2

O

Pyridoxal 5-phosphate (20 mol %), alcalase

OH NH2

2-Methyl-2-propanol/ H2O (19 : 1) pH = 8.5, 40 oC, 4 h

92% yield, 98% ee

Figure 4.26 DKR of amino acids catalyzed by pyridoxal 5-phosphate.

j105

j 4 Dynamic Kinetic Resolutions

106

O

O O NH2

CH3

Pyridoxal, CALB, NH 3

NH2

t-Butyl alcohol/ t-Butyl methyl ether (70 : 30 v/v) o

NH2 85% yield, 88% ee

–20 C , 66 h

Figure 4.27 DKR of amino acid derivatives.

CAL-A, Et3N

CO2Me

N H

CO2Me

N

O Pr

O t-Butyl methyl ether, 25 oC

O

86% yield, 97% ee

1h Figure 4.28 DKR of -amino esters catalyzed by CAL-A and acetaldehyde released in situ.

4.7 Enzyme-Catalyzed Racemization

Asano et al. have developed an approach for the synthesis of D-amino acids through DKR using a two-enzyme system [55]. They had previously reported the discovery of new D-stereospeciﬁc hydrolases that can be applied to KR of racemic amino acid amides to yield D-amino acids. Combination of a D-stereospeciﬁc hydrolase with an amino acid amide racemase allows performing DKR of L-amino acid amides yielding enantiomerically pure D-amino acids in excellent yields (Figure 4.29). O H2N

O

D-aminopeptidase

Me NH2

α-Amino-ε-caprolactam racemase Potassium phosphate buffer, 30 oC

HO

Me NH2

7h 99.7% yield, >99% ee

Figure 4.29 DKR using a two-enzyme system.

4.8 Racemization through SN2 Displacement

Williams and coworkers have reported a DKR of a-bromo [56a] and a-chloro esters [56b]. In the latter case, the KR is catalyzed by commercially available cross-linked enzyme crystals derived from Candida cylindracea lipase. The racemization takes place through halide SN2 displacement. The DKR is possible because the racemization of the substrate is faster than that of the product (carboxylate). For the ester, the empty p*(C¼O) orbital is able to stabilize the SN2 transition state by accepting

4.9 Other Racemization Methods

O

O

Candida cylindracea lipase

OMe Cl

OH

ResinPPh 3Cl

Cl

phosphate buffer, pH = 7

90% conversion, 90% ee

Figure 4.30 Enzyme- and chloride-catalyzed DKR.

electron density. However, the carboxylate is more electron-rich and therefore less able to facilitate an SN2 reaction. Racemization is performed by the use of a chloride source. The best results was obtained with a resin-bound phosphonium chloride (Figure 4.30). Faber and coworkers have reported a DKR of mandelic acid by using a lipasecatalyzed O-acylation followed by a racemization catalyzed by mandelate racemase. However, these two transformations do not take place simultaneously in the same pot. When the sequence was repeated four times, (S)-O-acetylmandelic acid was obtained in 80% isolated yield and >98% ee [57].

4.9 Other Racemization Methods

In this section we will discuss some DKRs in which racemization occurs spontaneously during the enzymatic resolution, and without further addition of any reagent. 4.9.1 DKR of 5-Hydroxy-2-(5H)-Furanones

During the KR of this kind of substrates, racemization occurs spontaneously as a consequence of the labile stereogenic center at C-5 [58]. The interconversion occurs by mutarotation, allowing a DKR. The KR is catalyzed by a lipase in the presence of vinyl acetate (Figure 4.31). 4.9.2 DKR of Hemiaminals

Spontaneous racemization also occurs during the enzyme-catalyzed acetylation of hemiaminals [59]. It is thought that the racemization occurs owing to ring opening at

HO

O

O

Lipase PS

AcO

O

O

OAc Hexane 100% conversion, 83% ee

Figure 4.31 DKR in which racemization occurs spontaneously.

j107

j 4 Dynamic Kinetic Resolutions

108

OH

OAc

OH Lipase QL

NR

NR

NR

O

O

O

OAc 100% conversion, >99% ee

Hexane, 60 o C 1d O

NHR O Figure 4.32 DKR in which racemization occurs spontaneously.

the C-3 position under the reaction conditions (Figure 4.32). The KR is catalyzed by lipase QL (Alcaligenes sp.) in the presence of isopropenyl acetate.

4.9.3 DKR of 8-Amino-5,6,7,8-tetrahydroquinoline

Very recently Crawford has reported an interesting enzymatic transformation catalyzed by CALB in which the yield was >60% and the product was obtained in very high optical purity (Figure 4.33) [60]. Furthermore, the remaining starting material was racemic. A mechanistic study performed by the authors showed that CALB and EtOAc are required for racemization to take place. They claimed that enzyme-catalyzed oxidation of the amine substrate takes place, and a ketone intermediate is produced. The ketone reacts with the substrate to form an enamine. This process results in racemization of the substrate (Figure 4.34). It is not clear how the enzyme and EtOAc could dehydrogenate the amine. To conﬁrm this racemization mechanism, Crawford et al. added 5 mol % of the ketone to the reaction mixture and obtained the product in 78% yield in >98% ee. This DKR is therefore catalyzed by a carbonyl compound, and can be compared to those shown in Section 4.6.

CALB, EtOAc o

Toluene, 50 C

N NH2

2d

N NHAc > 60% yield, >95% ee

Figure 4.33 DKR in which racemization occurs spontaneously.

4.10 Concluding Remarks CALB, EtOAc o

N

Toluene, 50 C

N

N

N

N NH2

N NH

NH2

O

O

N HN

N NH2 rac

Figure 4.34 Mechanism of the racemization suggested in ref. 60.

4.10 Concluding Remarks

During the past few years, great achievements in enzyme-catalyzed DKR have been obtained. The efﬁciency and scope have improved dramatically. In this process, an enzyme-catalyzed resolution is combined with an in situ racemization allowing a 100% theoretical yield of a variety of enantiopure products to be obtained. The use of enzymes in organic solvents has contributed toward development of DKRs in which racemization is catalyzed by metal complexes. The DKR of alcohols can now be accomplished by the use of a variety of different metals under very mild reaction conditions. It can be applied to the preparation of (R)- or (S)-conﬁgured substrates by the use of lipases or proteases, respectively. Also, the acyl group introduced during the enzymatic resolution has been employed in further transformations, increasing the scope of DKR and the development of highly atom-economical processes. DKR has also been used for the synthesis of chiral polymers, which is impossible to accomplish without the presence of a racemization catalyst. Furthermore, DKR of amines can now be performed very efﬁciently by the combination of enzymes and metal catalysts. Many other racemization methods can be employed in combination with enzymes, such as base- or acid-catalyzed racemizations, aldehyde- or enzymecatalyzed racemizations, or even spontaneous racemizations that take place under the reaction conditions. In all cases, excellent yields and enantioselectivities are obtained, making enzyme-catalyzed DKRs a useful and efﬁcient strategy for the synthesis of enantiopure compounds. The next future will also witness many new contributions to this area. New racemization catalysts, which are more efﬁcient, environmentally friendly, and cheaper are being developed. Also, use of other enzymes will broaden the scope of this transformation. In particular, the new methods available today such as directed evolution, coupled with advances in high-throughput screening technologies, will provide new enzymes with the desired properties to perform new transformations.

j109

j 4 Dynamic Kinetic Resolutions

110

References 1 (a) Halpern, J. and Trost, B.M. (2004) Proceedings of the National Academy of Sciences of the United States of America, 101, 5347. (b) Ojima, I. (ed.) (2000) Catalytic Asymmetric Synthesis, 2nd edn, Wiley-VCH Verlag GmbH, New York. (c) Jacobsen, E.N., Pfaltz, A. and Yamamoto, H. (eds) (1999) Comprehensive Asymmetric Catalysis, Springer, Berlin. (d) Noyori, R. (1994) Asymmetric Catalysis in Organic Synthesis, John Wiley & Sons, Ltd, New York. (e) Drauz, K. and Waldmann, H. (eds) (2002) Enzyme Catalysis in Organic Synthesis: A Comprehensive Handbook, 2nd edn, Wiley-VCH Verlag GmbH, Weinheim, Vols. I–III. (f) Faber, K. (2000) Biotransformations in Organic Chemistry, 4th edn, Springer, Berlin. (g) Dalko, P.I. and Moisan, L. (2004) Angewandte Chemie International Edition, 43, 5138–5175. 2 Knowles, W.S., Sabacky, M.J., Vineyard, B.D. and Weinkauff, D.J. (1975) Journal of the American Chemical Society, 97, 2567–2568. 3 Ohkuma, T. Kitamaura, M. and Noyori, R. (2000) Asymmetric hydrogenations, in Catalytic Asymmetric Synthesis (ed. I. Ojima), 2nd edn, Wiley-VCH Verlag GmbH, New York. 4 Breuer, M., Ditrich, K., Habicher, T., Hauer, B., Keeler, M., St€ urmer, R. and Zelinski, T. (2004) Angewandte Chemie International Edition, 43, 788–824. 5 Eliel, E.L., Wilen, S.H. and Mander, L.N. (1994) Stereochemistry of Organic Compounds, John Wiley & Sons, Ltd, New York. 6 Kagan, H.B. and Fiaud, J.C. (1988) Topics in Stereochemistry, 18, 249–330. 7 (a) Sih, C.J. and Wu, S.-H. (1989) Topics in Stereochemistry, 19, 63–125. (b) Chen, C.-S., Wu, S.-H., Girdaukas, G. and Sih, C.J. (1987) Journal of the American Chemical Society, 109, 2812–2817. (c) Chen, C.-S., Fujimoto, Y., Girdaukas, G. and Sih, C.J. (1982) Journal of the American Chemical Society, 104, 7294–7299

8 The selectivity factor used in nonenzymatic kinetic resolutions is known as s, see: (a) reference [6]. (b) Martin, V.S., Woodard, S.S., Katsuki, T., Yamada, Y., Ikeda, M. and Sharpless, K.B. (1981) Journal of the American Chemical Society, 103, 6237–6240 9 Bornscheuer, U.T. and Kazlauskas, R.J. (1999) Hydrolases in Organic Synthesis, Wiley-VCH Verlag GmbH, Weinheim. 10 (a) Klibanov, A.M. (2001) Nature, 409, 241–246. (b) Halling, P.J. (2000) Current Opinion in Chemical Biology, 4, 74–80. (c) Zaks, A. and Klibanov, A.M. (1985) Proceedings of the National Academy of Sciences of the United States of America, 82, 3192–3196. 11 Koh, J.H., Jeong, H.M. and Park, J. (1998) Tetrahedron Letters, 39, 5545–5548. 12 St€ urmer, R. (1997) Angewandte Chemie International Edition, 36, 1173–1174. 13 Allen, J.V. and Williams, J.M.J. (1996) Tetrahedron Letters, 37, 1859–1862. 14 Choi, K.L., Suh, J.H., Lee, D., Lim, I.T., Jung, J.Y. and Kim, M.-J. (1999) Journal of Organic Chemistry, 64, 8423–8424. 15 Akai, S., Tanimoto, K., Kanao, Y., Egi, M., Yamamoto, T. and Kita, Y. (2006) Angewandte Chemie International Edition, 45, 2592–2595. 16 Martın-Matute, B., Edin, M., Bogar, K., Kaynak, F.B. and B€ackvall, J.-E. (2005) Journal of the American Chemical Society, 127, 8817–8825. 17 (a) Clapham, S.E., Hadzovic, A. and Morris, R.H. (2004) Coordination Chemistry Reviews, 248, 2201–2237 (b) Gladiali, S. and Alberico, E. (2004) In Transition Metals for Organic Synthesis (eds M. Beller and C. Bolm), 2nd edn, WileyVCH Verlag GmbH, Weinheim, Vol. 2, pp. 145–166 (c) B€ackvall, J.-E. (2002) Journal of Organometallic Chemistry, 652, 105–111. (d) Huerta, F.F., Minidis, A.B.E. and B€ackvall, J.E. (2001) Chemical Society Reviews, 30, 321–331. (e) Wills, M., Palmer, M., Smith, A., Kenny, J. and Walsgrove, T.

References

18

19

20

21 22

23

24

25

26

(2000) Molecules, 5, 4–18. (f) Palmer, M. and Wills, M. (1999) Tetrahedron: Asymmetry, 10, 2045–2061. (g) MartınMatute, B., Åberg, J.B., Edin, M. and B€ackvall, J.E. Racemization of Secondary Alcohols Catalyzed by Cyclopentadienyl ruthenium Complexes. Insight into the Reaction Mechanism. Chem. Eur. J. 2007, 13, 6063–6072. (a) Pamies, O. and B€ackvall, J.-E. (2001) Chemistry-A European Journal, 7, 5052–5058. (b) Samec, J.S.M., B€ackvall, J.-E., Andersson, P.G. and Brandt, P. (2006) Chemical Society Reviews, 35, 237–248. Dinh, P.M., Howarth, J.A., Hudnott, A.R., Williams, J.M.J. and Harris, W. (1996) Tetrahedron Letters, 37, 7623–7626. (a) Larsson, A.L.E., Persson, B.A. and B€ackvall, J.-E. (1997) Angewandte Chemie International Edition in English, 36, 1211–1212. (b) Persson, B.A., Larsson, A.L.E., Le Ray, M. and B€ackvall, J.-E. (1999) Journal of the American Chemical Society, 121, 1645–1650. Menashe, N. and Shvo, Y. (1991) Organometallics, 10, 3885–3891. Verzijl, G.K.M., De Vries, J.G. and Broxterman, Q.B. (2003) PCT International Application WO 0190396 Al 20011129. Process for preparation of enantiomerically enriched esters and alcohols. (a) Choi, J.E., Kim, Y.H., Nam, S.H., Shin, S.T., Kim, M.J. and Park, J. (2002) Angewandte Chemie International Edition, 41, 2373–2376. (b) Choi, J.H., Choi, Y.K., Kim, Y.H., Park, E.S., Kim, E.J., Kim, M.-J. and Park, J. (2004) Journal of Organic Chemistry, 69, 1972–1977. Martın-Matute, B., Edin, M., Bogar, K. and B€ackvall, J.-E. (2004) Angewandte Chemie International Edition, 43, 6535–6539 DYKAT of 1,4-symmetrical diols: Martın-Matute, B., Edin, M. and B€ackvall, J.-E. (2006) Chemistry-A European Journal, 12, 6053–6061. (a) DYKAT of 1,2-diols: Edin, M., Martın-Matute, B. and B€ackvall, J.E. (2006)

27

28

29

30

31

32

33

34

35 36

37

38 39

Tetrahedron: Asymmetry, 17, 708–715. (b) DYKAT of 3,5-piperidinediols: Olofsson, B., Bogar, K., Fransson, A.B.L. and B€ackvall, J.E. (2006) Journal of Organic Chemistry, 71, 8256–8260. Kim, N., Ko, S.-B., Kwon, M.S., Kim, M.-J. and Park, J. (2005) Organic Letters, 7, 4523–4526. Dijksman, A., Elzinga, J.M., Li, Y.-X., Arends, I.W.C.E. and Sheldon, R.A. (2002) Tetrahedron: Asymmetry, 13, 879–884. Kazlauskas, R.J., Weissﬂoch, A.N.E., Rappaport, A.T. and Cuccia, L.A. (1991) Journal of Organic Chemistry, 56, 2656–2665. S-selective DKR: (a) Kim, M.J., Chung, Y., Choi, Y., Lee, H., Kim, D. and Park, J. (2003) Journal of the American Chemical Society, 125, 11494–11495. (b) Boren, L., Martın-Matute, B., Xu, Y., Córdova, A. and B€ackvall, J.-E. (2006) Chemistry-A European Journal, 12, 225–232. Berkessel, A., Sebastian-Ibarz, M.L. and M€ uller, T.N. (2006) Angewandte Chemie International Edition, 45, 6567–6570. Akai, S., Tanimoto, K. and Kita, Y. (2004) Angewandte Chemie International Edition, 43, 1407–1410. van As, B.A.C., van Buijtenen, J., Heise, A., Broxterman, Q.B., Verzjil, G.K.M., Palmans, A.R.A. and Meijer, E.W. (2005) Journal of the American Chemical Society, 127, 9964–9965. Hilker, I., Rabani, G., Verzijl, G.K.M., Palmans, A.R.A. and Heise, A. (2006) Angewandte Chemie International Edition, 45, 2130–2132. Reetz, M.T. and Schimossek, K. (1996) Chimia, 50, 668–669. Paetzold, J. and B€ackvall, J.-E. (2005) Journal of the American Chemical Society, 127, 17620–17621. Parvulesc, A., De Vos, D. and Jacobs, P. (2005) Chemical Communications, 5307–5309. Stirling, M., Blacker, J. and Page, M.I. (2007) Tetrahedron Letters, 48, 1247–1250. (a) Tan, D.S., G€ unter, M.M. and Drueckhammer, D.G. (1995) Journal of the

j111

j 4 Dynamic Kinetic Resolutions

112

40

41

42

43

44

45

46

47

American Chemical Society, 117, 9093–9094. (b) Um, P.-J. and Drueckhammer, D.G. (1998) Journal of the American Chemical Society, 120, 5605–5610. Wen, W.-Y., Ng, I.-S. and Tsai, S.-W. (2006) Journal of Chemical Technology and Biotechnology, 81, 1715–1721. (a) Turner, N.J., Winterman, J.R., McCague, R., Parrat, J.S. and Taylor, S.J.C. (1995) Tetrahedron Letters, 36, 1113–1116. (b) Brown, S.A., Parker, M.-C. and Turner, N.J. (2000) Tetrahedron: Asymmetry, 11, 1687–1690. See also: (c) Bevinakatti, H.S. Newadkar, R.V. and Banerji, A.A. (1990) Journal of the Chemical Society, Chemical Communications, 1091–1092. (d) Bevinakatti, H.S., Banerji, A.A., Newadkar, R.V. and Mokashi, A. (1992) Tetrahedron: Asymmetry, 3, 1505–1508. (e) Gu, R.-L., Lee, I.S. and Sih, C.J. (1992) Tetrahedron Letters, 33, 1953–1956. (f) Crich, J., Brieva, R., Marquart, P., Gu, R.-L., Flemming, S. and Sih, C.J. (1993) Journal of Organic Chemistry, 58, 3252–3258. (a) Suzuki, M., Yamazaki, T., Ohta, H., Shima, K., Ohi, K., Nishiyama, S. and Sugai, T. (2000) Synlett, 189–192. (b) See also: Lo, H.-H., Kao, C.-H., Lee, D.-S., Yang, T.-K. and Hsu, W.-H. (2003) Chirality, 15, 699–702. Taniguchi, T. and Ogasawara, K. (1997) Chemical Communications, 1399–1400. (a) Wuyts, S., De Temmerman, D., De Vos, D. and Jacobs, P.A. (2003) Chemical Communications, 1928–1929. (b) Wuyts, S., De Temmerman, K., De Vos, D.E. and Jacobs, P.A. (2005) Chemistry-A European Journal, 11, 386–397. Ödman, P., Wessjohann, L.A. and Bornscheuer, U.T. (2005) Journal of Organic Chemistry, 70, 9551–9555. Inagaki, M., Hiratake, J., Nishioka, T. and Oda, J. (1992) Journal of Organic Chemistry, 57, 5643–5649. (a) Paizs, C., Toxa, M., Majdik, C., T€ahinen, P., Irimie, F.-D. and Kanerva, L.T. (2003) Tetrahedron: Asymmetry, 14, 619–627. (b)

48

49 50

51

52

53

54 55

56

Paizs, C., T€ahinen, P., Lundell, K., Poppe, L., Irimie, F.-D. and Kanerva, L.T. (2003) Tetrahedron: Asymmetry, 14, 1895–1904. (c) Paizs, C., T€ahinen, P., Toxa, M., Majdik, C., Irimie, F.-D. and Kanerva, L.T. (2004) Tetrahedron, 60, 10533–10540. (d) Veum, L., Kanerva, L.T., Halling, P.J., Maschmeyer, T. and Hanefeld, U. (2005) Advanced Synthesis and Catalysis, 347, 1015–1021. Desantis, G., Zhu, Z., Greenberg, W.A., Wong, K., Chaplin, J., Hanson, S.R., Farwell, B., Nicholson, L.W., Rand, C.L., Weiner, D.P., Robertson, D.E. and Burk, M.J. (2002) Journal of the American Chemical Society, 124, 9024–9025. Brand, S., Jones, M.F. and Rayner, C.M. (1995) Tetrahedron Letters, 36, 8493–8496. Yamada, S., Hongo, C., Yoshioka, R. and Chibata, I. (1982) Journal of Organic Chemistry, 48, 843–846. Chen, S.-T., Huang, W.-H. and Wang, D.-T. (1994) Journal of Organic Chemistry, 59, 7580–7581. Parmar, V.S., Singh, A., Bisht, K.S., Kumar, N., Belokon, Y.N., Kochetkov, K.A., Iknooikov, N.S., Orlova, S.A., Tararov, V.I. and Saveleva, T.F. (1996) Journal of Organic Chemistry, 61, 1223–1227. (a) Hacking, M.A.P.J., Wegman, M.A., Rops, J., van Rantwijk, F. and Sheldon, R.A. (1998) Journal of Molecular Catalysis B: Enzymatic, 5, 155–157. (b) Wegman, M.A., Hacking, M.A.P.J., Rops, J., Pereira, P., Van Rantwijk, F. and Sheldon, R.A. (1999) Tetrahedron: Asymmetry, 10, 1739–1750. Liljeblad, A., Kiviniemi, A. and Kanerva, L.T. (2004) Tetrahedron, 60, 671–677. (a) Asano, Y. and Yamaguchi, S. (2005) Journal of the American Chemical Society, 127, 7696–7697. (b) See also: May, O. Verseck, S. Bommarius, A. and Drauz, D. (2002) Organic Process Research and Development, 6, 452–457. (a) Jones, M.M. and Williams, J.M.J. (1998) Chemical Communications, 2519–2520. (b) Haughton, L. and Williams, J.M.J. (2001) Synthesis, 943–946.

References 57 Strauss, U.T. and Faber, K. (1999) Tetrahedron: Asymmetry, 10, 4079–4081. 58 Thuring, J.W.J.F., Klunder, A.J.H., Nefkens, G.H.L., Wegman, M.A. and Zwanenburg, B. (1996) Tetrahedron Letters, 37, 4759–4760.

59 Sharfuddin, M., Narumi, A., Iwai, Y., Miyazawa, K., Yamada, S., Kakuchi, t. and Kaga, H. (2003) Tetrahedron: Asymmetry, 14, 1581–1885. 60 Crawford, J.B., Skerlj, R.T. and Bidger, G.J. (2007) Journal of Organic Chemistry, 72, 669–671.

j113

j115

5 Deracemization and Enantioconvergent Processes Nicholas J.Turner

5.1 Introduction

A dominant theme in modern asymmetric catalysis is the development of processes that involve the total conversion, rather than resolution, of both enantiomers of a racemate into a single enantiomer of the product. Such transformations of racemates become equivalent to asymmetric transformations on prochiral substrates and hence are highly desirable, not least because a racemate is often a convenient starting point for the synthesis of a molecule possessing one or more stereogenic centers. Typically the key functional group that is attached to the stereogenic center will be an alcohol, carboxylic acid/amide, amine, nitrile, epoxide, or thiol and hence biocatalytic methods are continually being sought and developed to manipulate these groups in novel ways [1]. Three distinct methods have been developed for achieving the total conversion of a racemate in high yield and enantiomeric excess (Figure 5.1). The ﬁrst and by far the most popular approach is dynamic kinetic resolution (DKR), which is dealt with in Chapter 2.1 by Jan-Erling Backvall. In a typical DKR, an enantioselective enzyme is used to transform a racemic mixture in which the two enantiomers of the substrate undergo rapid interconversion via racemization. The present chapter deals with two alternative strategies, the ﬁrst of which is called deracemization. Deracemization is deﬁned here and elsewhere [2] as a reaction whereby two enantiomers are interconverted by a stereoinversion mechanism, allowing the racemate to be transformed into a single enantiomer without any net change in the constitution of the molecule, that is, there is no product, but simply an optical enrichment of the substrate. The second approach is based upon enantioconvergent transformations in which a necessary condition is that one enantiomer is transformed with inversion of conﬁguration at the stereogenic center. A number of previous reviews [2–6] have dealt with both deracemization and enantioconvergent processes and hence this chapter will focus primarily on the recent literature.

Asymmetric Organic Synthesis with Enzymes. Edited by Vicente Gotor, Ignacio Alfonso, and Eduardo Garcıa-Urdiales Copyright 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 978-3-527-31825-4

j 5 Deracemization and Enantioconvergent Processes

116

A

P

A [X]

Racemization + (a)

B

Q

A

B

(b)

Retention

+ (c)

P +

B Inversion Q

Figure 5.1 Schematic illustration of (a) dynamic kinetic resolution, (b) deracemization, and (c) enantioconvergent processes.

5.2 Deracemization Processes 5.2.1 Cyclic Oxidation–Reduction Systems

One general type of deracemization process depends upon establishing a catalytic cycle involving the combination of (i) an enzyme-catalyzed enantioselective transformation (oxidation) of a racemic mixture into a stable prochiral intermediate and (ii) the reduction of this intermediate back to the starting material in a separate step (Figure 5.2). Provided that the enzyme is enantioselective, the faster reacting enantiomer will be rapidly depleted, and following a number of cycles, the slower reacting enantiomer will accumulate in the reaction. The ﬁnal enantiomeric excess will be determined directly by the enantioselectivity (kR/kS) of the enzyme-catalyzed reaction. Assuming that the enzymatic reaction is highly enantioselective, then even after only four cycles the enantiomeric excess will have reached 93.4% whereas after seven catalytic cycles the enantiomeric excess is >99% (Figure 5.3). This type of deracemization is really a stereoinversion process in that the reactive enantiomer undergoes stereoinversion during the process. One of the challenges of developing this type of process is to ﬁnd conditions under which the enzyme catalyst and chemical reactant can coexist, particularly in the case of redox chemistry in which the coexistence of an oxidant and reductant in the same reaction vessel is difﬁcult to achieve. For this

SR

k red

SR, SS = Substrate enantiomers

kR I

I = Achiral/prochiral intermediate

kS SS

k R , k S, k red = Rate constants

k red k R»k S

Figure 5.2 Cyclic deracemization process involving sequential enzyme-catalyzed oxidation and nonenzymatic reduction.

% Enantiomer

5.2 Deracemization Processes

100 90 80 70 60 50 40 30 20 10 0 0

1

2

3

L-Amino

acid

D-Amino

acid

j117

4

Number of cycles Figure 5.3 Progression of an ideal cyclic deracemization process.

reason, perhaps it is not surprising that the only known examples of this type of cyclic oxidation–reduction system that have been reported involve an enzyme as the catalyst for the oxidation step. The ﬁrst experimental demonstration of this type of deracemization was provided by Hafner and Wellner [7] who used D-amino acid oxidase in combination with sodium borohydride to generate a small quantity (about 5% conversion) of L-alanine from D-alanine (Figure 5.4). In this process, the D-alanine (1: R ¼ CH3) is initially oxidized to the corresponding imine (2) by the D-amino acid oxidase followed by concomitant reduction of the imine nonselectively, which generates a small quantity of L-enantiomer. Interestingly, Hafner et al. estimated that the reduction of the imine intermediate was about 35· faster than hydrolysis to the corresponding a-keto acid (3). Subsequently, Soda et al. [8,9] recognized the possibility of applying this concept

NH 3+ R

Oxidation acid oxidase

D -Amino

CO2 – (D)-1

NH Reduction R [H]

CO2H 2

NH3 + R

O

+ H 2O

CO 2– (L)-1

Figure 5.4 Deracemization of a-amino acids using D-amino acid oxidase in combination with sodium borohydride.

+ R

CO2H 3

NH4+

j 5 Deracemization and Enantioconvergent Processes

118

to preparative deracemization reactions and starting with both D/L-proline and D/Lpipecolic acid were able to achieve high conversions (>95%) and high enantiomeric excesses (>99%) to the corresponding L-enantiomers using the same D-amino acid oxidase/sodium borohydride combination. The work of Soda highlighted two key questions. First, cyclic amino acids are more suitable for this type of deracemization process in that the oxidized imine is cyclic and hence more stable than the corresponding imine derived from an acyclic amino acid, which is prone to hydrolysis to the a-keto acid. Secondly, excesses of sodium borohydride were required to achieve complete deracemization as a result of the instability of sodium borohydride in water. Subsequently, Turner et al. demonstrated that acyclic, as well as cyclic, amino acids could be deracemized in this manner although the yields were slightly lower (about 80%) owing to competing hydrolysis. More importantly, they discovered that water-stable sodium cyanoborohydride could be used in place of sodium borohydride permitting a large reduction in the number of equivalents (about ﬁve) of reducing agent [10]. Thereafter, in collaboration with scientists at Great Lakes in the United States, both amine–borane complexes and catalytic transfer hydrogenation using palladium and ammonium formate were found to be effective methods for in situ reduction of the imine [11]. In addition, for the ﬁrst time D-amino acids were prepared in high yield and enantiomeric excess by use of the L-amino acid oxidase from Proteus myxofaciens. The catalytic transfer hydrogenation approach has been extensively developed by the company Ingenza (www.ingenza.com), which has now devised cost-effective, large-scale processes for the deracemization of speciﬁc amino acids using this approach [12–15]. As outlined above, the oxidation–reduction deracemization process is in essence a stereoinversion and hence could in principle be applied to substrates possessing more than one stereogenic center. Turner et al. demonstrated the interconversion of diasteroisomers of amino acids, for example, the conversion of L-isoleucine (4) to Dallo-isoleucine (5) in high yield and enantiomeric excess/diastereoisomeric excess (Figure 5.5) [16]. In a further illustration of this approach, they collaborated with Dowpharma (Chirotech), Cambridge, UK to combine catalytic asymmetric hydrogenation of dehydroamino acids, using chiral rhodium–DuPhos catalysts, with the oxidase stereoinversion technology to prepare a wide range of b-substituted phenylalanines (6) in high diastereoisomeric excess and enantiomeric excess (Figure 5.6) [17].

NH2 CO2 H Me L-isoleucine

NH 2

Proteus myxofaciens L-AAO

4

CO 2 H

[H]

Me D-allo-isoleucine

NH3 -BH 3; 87% yield; 99% d.e. Pd/C-HCO 2NH4 ; 61% yield; 96% d.e. Figure 5.5 Stereoinversion of L-isoleucine to D-allo-isoleucine.

5

5.2 Deracemization Processes

NHAc Ar

CO2 Me

ii. HCl

Me

D-AAO NH 3:BH 3

NH 2

i. Rh(R,R)-EtDuPhos H2 Ar

j119

NH2 Ar

CO2 H

CO2H

Me

Me 6 e.e.'s >98%

Figure 5.6 Synthesis of b-substituted phenylalanines.

Recently Turner and coworkers have sought to extend the deracemization method beyond a-amino acids to encompass chiral amines. Chiral amines are increasingly important building blocks for pharmaceutical compounds that are either in clinical development or currently licensed for use as drugs (Figure 5.7). At the outset of this work, it was known that type II monoamine oxidases were able to catalyze the oxidation of simple amines to imines in an analogous fashion to amino acid oxidases. However, monoamine oxidases generally possess narrow substrate speciﬁcity and moreover have been only documented to catalyze the oxidation of simple, nonchiral

Me

Me O

Me2 N

N

CO 2 H

O Et

N H

O

OEt

Me2 N

O

N

(S)-rivastigmine

(S)-repaglinide

(S)-dapoxetine(phaseII)

O Me

N HO 2 C

N

O

Cl

O

N N

Et Et

O

(R)-levocetirizine

N

N H O

(S)-DMP 777 (phase II) F2 HC O HN

N

N CO 2H

Ph

O O

N

O (R)-garenoxacin (phase III)

Figure 5.7 Drugs containing chiral amines.

(S)-solifenacin (phase III)

O O

j 5 Deracemization and Enantioconvergent Processes

120

NH 2

NH3 :BH 3

CH3 Fast

NH

(S)-α-methylbenzylamine 7 MAO-N (Asn336Ser) Slow

Prochiral imine intermediate 8

NH 2 CH3 NH3 :BH 3 (R)-α-methylbenzylamine 7 77% yield; 93% e.e.

Figure 5.8 Deracemization of a-methylbenzylamine by using a variant monoamine oxidase in combination with NH3: BH3.

amines. The major objective of this work therefore was to apply directed evolution methods to create amine oxidase variants possessing broad substrate speciﬁcity and high enantioselectivity. The initial goal was to identify variant enzymes that are able to carry out the deracemization of the model substrate a-methylbenzylamine (7) (Figure 5.8). Monoamine oxidase N (MAO-N) from Aspergillus niger was expressed in Escherichia coli and a library of variants was generated using the E. coli XL-1 Red mutator strain. The variants were screened for activity against a-methylbenzylamine as substrate on solid phase using a high-throughput colorimetric screen. Using this approach, Alexeeva et al. successfully identiﬁed a variant (Asn336Ser) that was able to enantioselectively oxidize (S)-a-methylbenzylamine (7) to the corresponding imine (8) [18]. By contrast, the parent wild-type enzyme had very low activity toward (S)-(7), although there was some evidence that it was enantioselective. Introduction of the point mutation resulted in an improvement of catalytic activity of about 47-fold and in combination with ammonia borane, this variant, amine oxidase, was used to deracemize amethylbenzylamine to give the (R) enantiomer in 77% yield and 93% ee. Subsequently Turner and coworkers were able to show that the Asn336Ser variant possessed broad substrate speciﬁcity, with the ability to oxidize a wide range of chiral amines of interest [19]. They also discovered a second mutation, Ile246Met, which conferred enhanced activity toward chiral secondary amines as exempliﬁed by the deracemization of racemic 1-methyltetrahydroisoquinoline (MTQ) (9) (Figure 5.9)[20]. By carrying out iterative rounds of mutagenesis and screening, it has been possible to identify further variants that have high activity toward a broad range of chiral amines including tertiary amines (10–13) (Figure 5.10). Interestingly, in this case it is now possible to enter the catalytic cycle from the precursor acyclic amino ketone rather than simply starting with the racemic tertiary amine. This approach is exempliﬁed by the conversion of ketone (14), which under aqueous conditions at

j121

5.2 Deracemization Processes

Me

Me (S)-amine oxidase variant

NH

NH NH 3:BH 3

(R/S)- 9

(R)-9 95% yield; 99% e.e.

Figure 5.9 Deracemization of MTQ 9.

pH 7.0 forms the iminium ion (15). In situ reduction of (15) with ammonia borane generates racemic nicotine (10), which is then subjected to the usual deracemization process producing the unnatural (R) enantiomer of nicotine in >95% yield and >95% ee. Recently, one of the evolved MAO-N variants has been shown to catalyze the enantioselective oxidation of O-methyl-N-hydroxylamines (e.g. 16) yielding the corresponding E-oxime (17) and recovered (R)-hydroxylamine (16) in enantiomerically pure form (Figure 5.11) [21]. Finally in this section on deracemization via cyclic oxidation/reduction methods, there has been some limited work carried out on the deracemization of secondary alcohols. Soda et al. [22] employed lactate oxidase in combination with sodium borohydride to deracemize D/L-lactate (18) via the intermediate pyruvate (19) (Figure 5.12).

H N

N Me

N 10

N

Me

H

H

MeO

Me

N

N

Me

Me

12

11

13 H N Me

N

(S)-10 NHMe

pH 7

N+

O

Me

N

N 14

(S)-amine oxidase variant NH3 :BH 3 H

15

N N

Me (R)- 10

Figure 5.10 Deracemization of tertiary amines.

j 5 Deracemization and Enantioconvergent Processes

122

OMe

HN

HN

MAO-N pH 7.4, 25 oC

Me

OMe

Me

N

OMe

Me

+

18h (R/ S)-16

(E )-17; 46%

(R)-16; 44%; 99% e.e.

Figure 5.11 Enantioselective oxidation of hydroxylamines.

OH H 3C

NaBH4 CO 2H

L-Lactate

L-Lactateoxidase

18

O H 3C

OH H 3C

CO 2 H

Pyruvate 19 NaBH4

CO 2H

D-Lactate

18

Figure 5.12 Deracemization of D/L-lactate (18).

In a related approach, Adam et al. used glycolate oxidase with D-lactate dehydrogenase for the deracemization of a wide range of racemic a-hydroxy acids (20) (Figure 5.13) [23]. 5.2.2 Microbial Deracemization of Secondary Alcohols Using a Single Microorganism

It is well known that certain microorganisms are able to effect the deracemization of racemic secondary alcohols with a high yield of enantiomerically enriched compounds. These deracemization processes often involve two different alcohol dehydrogenases with complementary enantiospeciﬁcity. In this context Porto et al. [24] have shown that various fungi, including Aspergillus terreus CCT 3320 and A. terreus CCT 4083, are able to deracemize ortho- and meta-ﬂuorophenyl-1-ethanol in good OH R

O

Glycolate oxidase CO 2H

OH

+ R

CO 2H

R

a-Hydroxyacid 20

D-Lactate Dehydrogenase/NADH

Figure 5.13 Deracemization of a-hydroxyacids (20).

CO 2 H

5.2 Deracemization Processes

O

OH Rhizopus oryzae pH = 7.5–8.5

O

(R/S)- 21

Rhizopus oryzae pH = 4.0–5.0

O

Rhizopus oryzae pH = 7.5–8.5 OH

OH (S)-21

(R)-21 Figure 5.14 Deracemization of benzoin (21).

yields and high enantiomeric excesses. Likewise Demir et al. [25] have reported the deracemization of racemic benzoin (21) using Rhizopus oryzae ATCC 9363 (Figure 5.14). Interestingly, they were able to use the pH of the medium to control the absolute conﬁguration of the enantiomer produced. Thus, at pH 7.5–8.5, the (R) enantiomer was obtained in 73–76% yield and 97% ee whereas at pH 4–5 the (S) enantiomer was produced (71% yield; 85% ee). Chadha et al., have published a series of papers on the deracemization of bhydroxyesters using whole cells of Candida parapsilosis. For example, deracemization of racemic ethyl 2-hydroxy-4-phenylbutanoic acid (22: R ¼ H) yielded the (S) enantiomer in 85–90% yield and >99% ee (Figure 5.15) [26]. Analogous deracemizations of racemic ethyl/methyl esters of mandelic acid similarly yielded the (S) products in high enantiomeric excess [27]. A number of para-substituted analogs were also investigated using immobilized cells of C. parapsilosis ATCC 7330 affording the corresponding (S) enantiomers in moderate yield and excellent optical purity [28]. Several substrates with different substitution patterns on the aromatic ring underwent the reaction, but with generally lower enantioselectivity. Studies using model substrates suggested that the sequential activities of an (R)-speciﬁc oxidase and an (S)-speciﬁc reductase were responsible for the biotransformation. Likewise, C. parapsilosis was investigated for substrate tolerance in the deracemization reactions with aryl a-hydroxy esters (23) (Figure 5.16) [29]. A range of OH

O

OH O

Candida parapsilosis Aqueous 25°C, 6 hr

R

O O

R

rac - 22 R = H, Me, Et, OMe, Cl, NO 2 Figure 5.15 Deracemization of b-hydroxyesters using Candida parapsilosis.

(S)- 22

j123

j 5 Deracemization and Enantioconvergent Processes

124

OH

OH O

R'

O Candida parapsilosis

O

R

R'

O

R (S)-23

rac-23 Figure 5.16 Deracemization of a-hydroxyesters (23).

substrates was studied to examine the electronic and steric effects of substituents on the aryl ring. When compared with nonsubstituted, para-Cl substitution resulted in a yield decrease from 75 to 65%, and only a small change in enantiomeric excess, from 99 to 98%. Yields of all substrates tested ranged from 52 to 93%, and enantiomeric excesses from 3 to 100%. Carnell et al. discovered that whole cells of Cunninghamella echinulata NRRL 1384 were able to deracemize racemic N-(1-hydroxy-1-phenylethyl)benzamide (24) to produce the (R) enantiomer (Figure 5.17) [30]. The deracemization involves fast, highly (S)-selective oxidation, followed by slower, partially (R)-selective reduction of the ketone (25). Optimization by removing competing extracellular amidase/protease activity resulted in 82% yield and 92% ee. 5.2.3 Deracemization of Alcohols Using Two Enzyme/Microorganism Systems

An alternative approach to the microbial deracemization of secondary alcohols is to use two different microorganisms with complementary stereoselectivity. Fantin et al. studied the stereoinversion of several secondary alcohols using the culture supernatants of two microorganisms, namely Bacillus stearothermophilus and Yarrowia lipolytica (Figure 5.18) [31]. The authors tested three main systems for deracemization. First, they used the supernatant from cultures of B. stearothermophilus, to which they added Y. lipolytica cells and the racemic alcohols. Secondly, they used the culture supernatant of Y. lipolytica and added B. stearothermophilus cells and the racemic alcohols. Finally, they resuspended the cells of both organisms in phosphate buffer and added the racemic alcohols. The best results were obtained in the ﬁrst system with 6-penten-2-ol (26) (100% ee and 100% yield). The phosphate buffer system gave OH

Cunninghamella echinulata NRRL 1384

Ph HN

O

O Ph

+

HN

Ph (±)-24

OH

O

Ph HN

Ph 25

O Ph

(R)-24 82%, 92% e.e.

Figure 5.17 Deracemization of alcohol (24) using Cunninghamella echinulata.

5.2 Deracemization Processes

OH Yarrowia lipolytica (S)-26

O

OH

Prochiral ketone

Bacillus stearothermophilus (R)-26

Figure 5.18 Stereoinversion of 6-hexen-2-ol (26) using two microorganism preparations.

100% ee and a respectable 91% yield, whereas the Y. lipolytica supernatant gave poor results (40% ee and 85% yield). Faber et al. have reported a novel process for the overall deracemization of racemic mandelic acid derivatives using a combination of an enantioselective lipase and a mandelate racemase activity from Lactobacillus paracasei (Figure 5.19) [32]. Using this approach, racemates of (27) were enantiomerically enriched using a lipase in organic solvent, followed by racemization of the unreacted enantiomer in buffer. Acylated derivatives (S)-(28) were obtained in yields >50% and >99% ee. Lipases with the opposite enantioselectivity produced (R)-28 in >99% ee. Subsequent chemical deacylation of (28) yielded enantiomerically enriched (27). Other racemization systems that may be amenable to conversion to deracemization processes in future have recently been reported by Faber and coworkers [33]. Resting cells of L. paracasei have been used for biocatalytic racemization of open-chain and cyclic dialkyl-, alkyl-aryl-, and diaryl-substituted acyloins (29/30) (Figure 5.20). Both Pseudomonas sp. lipase

OH R

CO2 H rac-27

Vinyl acetate, i-Pr 2 O, 25 ºC

OH R

OAc CO2H

+

R

(R)-27

CO2 H (S)-28

Lactobacillus paracasei pH 6.5, 42 ºC Figure 5.19 Deracemization of mandelic acid derivatives (27).

O

Lactobacillus paracasei DSM20207 OH

()n (S)- 29/30

Buffer pH 7 29: n = 1 30: n = 2

Figure 5.20 Racemization of cyclic acyloins.

O OH ()n (R)- 29/30

j125

j 5 Deracemization and Enantioconvergent Processes

126

O

OBn r ac-31

Rhodococcus sp. CBS 717.73 Buffer pH 8 iso-octane (5:1) 30 °C, 24 h

HO

O

OBn +

HO

(R)-31

OBn (R)-32

O

OBn (S)-31

72% yield; >97% e.e. Figure 5.21 Deracemization of epoxides using Rhodococcus sp.

cyclic acyloins (S)-(29) and (S)-(30) were racemized in a clean fashion with no side products detected. 5.2.4 Epoxides

An interesting system has been reported for the deracemization of racemic epoxides (31) (Figure 5.21) [34]. Rhodococcus sp. R312 was used for the enantiospeciﬁc hydrolysis of racemic epoxide (31) to the corresponding diol (R)-(32). The unreacted (R)-(31) in the product mixture was converted to (R)-(32) by acidcatalyzed epoxide ring opening, and subsequent chemical cyclization gave the epoxide (S)-(31) in high yield and enantiomeric excess. The enantioselectivity and conversion were also studied for several other whole cell systems and isolated enzymes under various conditions. 5.2.5 Carboxylic Acids

Ohta et al. have reported some elegant studies into the microbial deracemization of 2-aryl- (33) and 2-aryloxypropanoic acids using growing cells of Nocardia diaphanozonaria JCM 3208 (Figure 5.22) [35]. In both cases, the (R) enantiomer was preferentially obtained from the racemate. Although the reaction was found to be very sensitive to the structure of the substrate, in optimal cases, for example, for 2-(4-chlorophenoxy) propanoic acid, yields of up to 95% were obtained with an associated enantiomeric excess of 97%. Recently this group has reported a series of studies to elucidate the underlying mechanism of the deracemization reaction. Labeling studies, together with the use of speciﬁc inhibitors, suggest the participation of three enzymes, namely, an acyl-CoA synthetase, an arylpropionyl-CoA epimerase, and a hydrolase [36]. In a subsequent study using a resting cell system of N. diaphanozonaria JCM3208, deracemization of 4-substituted-2-thiopropanoic acids (34) were

5.3 Enantioconvergent Processes

CH 3

CH 3 acyl-CoA synthetase CO2 H

X

COSCoA

X Hydrolase

(R)-33 Net reaction deracemization

2-arylpropionylCoA epimerase

CH 3

CH3 X

CO2 H

X

Hydrolase

COSCoA

X

(S)-33 Figure 5.22 Deracemization of 2-aryloxypropanoic acids (33).

X

CH 3 S

Nocardia diaphanozonaria resting cells

CO2 H

X

48 h

CH 3 S

CO2 H

34 X = H, Cl Figure 5.23 Deracemization of 4-substituted-2-thiopropanoic acids.

examined (Figure 5.23) [37]. It was found that side oxidation reactions occurred when deracemizations were performed under aerobic conditions. Operation under an atmosphere of inert argon was found to promote the deracemization reactions giving good-to-excellent enantiomeric excesses/yields. Evidence was presented that deracemization is a competitive reaction against the b-oxidation pathway of fatty acids.

5.3 Enantioconvergent Processes

As outlined above, enantioconvergent processes require two separate reaction pathways in order to transform a racemic substrate into a single product enantiomer. This is accomplished by employing a catalyst, which transforms one of the substrate enantiomers to the product with retention of conﬁguration. Concurrently, another catalyst, with opposite enantioselectivity and opposite regioselectivity, transforms the other substrate enantiomer with inversion of conﬁguration (Figure 5.24).

j127

j 5 Deracemization and Enantioconvergent Processes

128

kR

SR

PR

SR, SS = Substrate enantiomers PR = Nonracemic product

kS SS

kR, kS = Rate constants

Figure 5.24 Principle of an enantioconvergent process.

5.3.1 Epoxide Hydrolysis

Xu et al. employed crude mung bean (Phaseolus radiatus L.) powder for the enantioconvergent hydrolysis of styrene oxides, particularly p-nitrostyrene oxide (35; R ¼ NO2) (Figure 5.25) [38]. The corresponding diol products (36) are important intermediates for the synthesis of dopamine receptor antagonists. The crude extract employed in this process contains two epoxide hydrolases, mung bean epoxide hydrolase A and B (mbEH A and B), neither of which is enantioselective, but both of which show identical regioselectivities with coefﬁcients greater than 90%. That is, both enzymes convert the (R) epoxide with retention of conﬁguration while converting the (S) epoxide with inversion of conﬁguration. Using this crude mung bean preparation, the production of (R)-p-nitrostyrene oxide was achieved in 82% ee and 83.5% yield. After recrystallization, greater than 99% ee was achieved. Furstoss et al. have reported their studies on the use of an epoxide hydrolase with four styrene oxide derivatives (Figure 5.26) [39]. The (R)-diol (43) was obtained in 91% ee at 100% conversion from racemic (42), demonstrating an enantioconvergent HO

O

OH

Mung bean extract pH 6.5, 40 oC

X

X (R)-36

35 X = H, NO2, Cl

O

77.4 – 88.4% 60.7 – 82.4% e.e. O

O

O O O HO2C

CO2H 40

37

38

39

O 41

Figure 5.25 Enantioconvergent hydrolysis of epoxides (35) to the corresponding diols (36) using mung bean epoxide hydrolase.

5.3 Enantioconvergent Processes

OH O

OH

Solanum tuberosum epoxide hydrolase H2O/1 % DMSO

Cl

Cl (R)-43

(rac)-42

Figure 5.26 Enantioconvergent hydrolysis of racemic epoxide (42).

process. A preparative scale reaction, using 3 g of 42 resulted in an enantiomeric excess of 88% of (R)-(43). 5.3.2 Sulfatases

Faber and coworkers have recently investigated the potential for using enantioselective alkyl sulfatases for the enantioconvergent hydrolysis of racemic sulfates to the corresponding alcohols [40]. A subset of known sulfatases are able to enantioselectively hydrolyze racemic alkyl sulfates to the corresponding alcohols with inversion of conﬁguration, thereby fulﬁlling a necessary condition of an enantioconvergent process. For example, they studied the alkylsulfatase activity of Rhodococcus ruber DSM 44541. Biotransformation of the racemic sulfate ester (44) gave enantiomerically pure (S)-conﬁgured alcohol (45) and unreacted sulfate ester (44), also of (S) conﬁguration (Figure 5.27) [41]. In a subsequent report [42], they used an inverting sulfatase from Rhodococcus sp. RS2 for initial resolution followed by treatment of the sulfate/alcohol mixture with aqueous tert-butyl methyl ether and dioxane in the presence of p-toluenesulfonic acid as a catalyst to effect hydrolysis of the remaining unreacted sulfate ester with retention of conﬁguration. The overall process yielded a single stereoisomeric alcohol in an enantioconvergent process. Finally Faber et al. also reported the screening of a range of hyperthermophilic sulfur-metabolizers for the ability to convert racemic sulfate esters (46a) and (46b) to the corresponding alcohols (47a) and (47b) (Figure 5.28). Sulfolobus acidocaldarius DSM 639 gave the most promising results and, under optimized conditions, hydrolysis of (R)-(46a) proceeded with inversion of conﬁguration to give (S)-(47a) in moderate-to-high yield and excellent enantiomeric excess. Several other substrates were also tested.

OSO3 C5 H11

C2 H5 44

Rhodococcus r uber DSM 44541 Tris-buffer pH 7.5

H

OSO3

C 5H 11

C 2H 5 (S)-44

H C5 H11

OH 2

C2 H5

(S)-45 e.e. = 99%

Figure 5.27 Enantioselective hydrolysis of sulfate esters.

SO4

j129

j 5 Deracemization and Enantioconvergent Processes

130

OSO 3- Na + R

Me

Sulfolobus acidocaldarius DSM 639 Culture medium

rac-46a: R = (CH 2) 5CH 3 rac-46b: R = CH2 Ph

OSO 3-Na+

OH R

Me

(S)-47a 32%, >99% e.e. (S)-47b 42%, >99% e.e.

Me

R

(S)-46a (S)-46b

Figure 5.28 Enantioselective hydrolysis of sulfate esters using S. acidocaldarius.

5.4 Conclusions and Future Prospects

Enzyme-catalyzed deracemization and enantioconvergent processes have recently emerged as attractive approaches for the efﬁcient preparation of chiral building blocks in enantiomerically enriched form. Together with enzyme-mediated DKRs, an increasing toolbox of biocatalytic technologies is now available to complement existing methods based upon kinetic resolution of a racemate or asymmetric transformations of a prochiral substrate. Current developments in enzyme discovery/enzyme evolution together with advances in biocatalyst formulation and rapid process development will signiﬁcantly impact this area in a short-to-medium term to provide an even wider portfolio of biocatalysts for practical application in academic and industrial laboratories.

References 1 Patel, R.N. (2006) Current Opinion in Drug Discovery and Development, 9, 741. 2 Gadler, P., Glueck, S.M., Kroutil, W., Nestl, B.M., Larissegger-schnell, B., Ueberbacher, B.T., Wallner, S.R. and Faber, K. (2006) Biochemical Society Transactions, 34, 296. 3 Gruber, C.C., Lavandera, I., Faber, K. and Kroutil, W. (2006) Advanced Synthesis and Catalysis, 348, 1789. 4 Turner, N.J. (2004) Current Opinion in Chemical Biology, 8, 114. 5 Kroutil, W. and Faber, K. (1998) Tetrahedron: Asymmetry, 9, 2901. 6 Stecher, H. and Faber, K. (1997) Synthesis, 1, 1. 7 Hafner, E.W. and Wellner, D. (1971) Proceedings National Academy of Sciences, 68, 987.

8 Huh, J.W., Yokoigawa, K., Esaki, N. and Soda, K. (1992) Journal of Fermentation and Bioengineering, 74, 189. 9 Huh, J.W., Yokoigawa, K., Esaki, N. and Soda, K. (1992) Bioscience Biotechnology and Biochemistry, 56, 2081. 10 Beard, T. and Turner, N.J. (2002) Chemical Communications, 246. 11 Alexandre, F.-R., Pantaleone, D.P., Taylor, P.P., Fotheringham, I.G., Ager, D.J. and Turner, N.J. (2002) Tetrahedron Letters, 43, 707. 12 Turner, N.J., Fotheringham, I. and Speight, R. (2004) Innovations in Pharmaceutical Technology, 4, 114. 13 Fotheringham, I. (2005) Speciality Chemicals Magazine, 25, 32.

Further Reading 14 Archer, I., Carr, R., Fotheringham, I., Speight, R. and Turner, N.J. (2005) Chimica E L Industria, 87, 100. 15 Fotheringham, I., Archer, I., Carr, R., Speight, R. and Turner, N.J. (2006) Biochemical Society Transactions, 34, 287. 16 Enright, A., Alexandre, F.-R., Roff, G., Fotheringham, I.G., Dawson, M.J. and Turner, N.J. (2003) Chemical Communications, 2636. 17 Roff, G.J., Lloyd, R.C. and Turner, N.J. (2004) Journal of the American Chemical Society, 126, 4098. 18 Alexeeva, M., Enright, A., Dawson, M.J., Mahmoudian, M. and Turner, N.J. (2002) Angewandte Chemie International Edition, 41, 3177. 19 Carr, R., Alexeeva, M., Enright, A., Eve, T.S.C., Dawson, M.J. and Turner, N.J. (2003) Angewandte Chemie International Edition, 42, 4807. 20 Carr, R., Alexeeva, M., Dawson, M.J., Gotor-Fernandez, V., Humphrey, C.E. and Turner, N.J. (2005) ChemBioChem, 6, 637. 21 Eve, T.S.C., Wells, A.S. and Turner, N.J. (2007) Chemical Communications, in press. 22 Oikawa, T., Mukoyama, S. and Soda, K. (2001) Biotechnology and Bioengineering, 73, 80. 23 Adam, W., Lazarus, M., Saha-M€oller, C.R. and Schreier, Peter (1998) Tetrahedron: Asymmetry, 9, 351. 24 Comasseto, J.V., Omori, Á.T., Andrade, L.H. and Porto, A.L.M. (2003) Tetrahedron: Asymmetry, 14, 711. 25 Demir, A.S., Hamamci, H., Sesenoglu, O., Neslihanoglu, R., Asikoglu, B. and Capanoglu, D. (2002) Tetrahedron Letters, 43, 6447. 26 Chadha, A. and Baskar, B. (2002) Tetrahedron: Asymmetry, 13, 1461.

27 Padhi, S.K., Titu, D., Pandian, N.G. and Chadha, A. (2006) Tetrahedron, 62, 5133. 28 Baskar, B., Pandian, N.G., Priya, K. and Chadha, A. (2005) Tetrahedron, 61, 12296. 29 Padhi, S.K. and Chadha, A. (2005) Tetrahedron: Asymmetry, 16, 2790. 30 Cardus, G.J., Carnell, A.J., Trauthwein, H. and Riermeier, T. (2004) Tetrahedron: Asymmetry, 15, 239. 31 Fantin, G., Fogagnolo, M., Giovannini, P.P., Medici, A. and Pedrini, P. (1995) Tetrahedron: Asymmetry, 6, 3047. 32 Larissegger-Schnell, B., Glueck, S.M., Kroutil, W. and Faber, K. (2006) Tetrahedron, 62, 2912. 33 Nestl, B.M., Kroutil, W. and Faber, K. (2006) Advanced Synthesis and Catalysis, 348, 873. 34 Simeó, Y. and Faber, K. (2006) Tetrahedron: Asymmetry, 17, 402. 35 Kato, D., Mitsuda, S. and Ohta, H. (2002) Organic Letters, 4, 371. 36 Kato, D., Mitsuda, S. and Ohta, H. (2003) The Journal of Organic Chemistry, 68, 7234. 37 Kato, D., Miyamoto, K. and Ohta, H. (2004) Tetrahedron: Asymmetry, 15, 2965. 38 Xu, W., Xu, J.H., Pan, J., Gu, Q. and Wu, X.Y. (2006) Organic Letters, 8, 1737. 39 Monterde, M.I., Lombard, M., Archelas, A., Cronin, A., Arand, M. and Furstoss, R. (2004) Tetrahedron: Asymmetry, 15, 2801. 40 Gadler, P. and Faber, K. (2007) TIBTECH, 25, 83. 41 Pogorevc, M., Kroutil, W., Wallner, S.R. and Faber, K. (2002) Angewandte Chemie International Edition, 41, 4052. 42 Wallner, S.R., Pogorevc, M., Trauthwein, H. and Faber, K. (2004) Engineering Life Sciences, 4, 512.

Further Reading Dunsmore, C.J., Carr, R., Fleming, T. and Turner, N.J. (2006) Journal of the American Chemical Society, 128, 2224.

Wallner, S.R. Nestl, B.M. and Faber, K. (2005) Organic and Biomolecular Chemistry, 3, 2652.

j131

j133

6 Transesterification and Hydrolysis of Carboxylic Acid Derivatives, Alcohols, and Epoxides Robert Ch^enevert, Pierre Morin, and Nicholas Pelchat

6.1 Introduction and General Aspects 6.1.1 Scope

This chapter presents a review of recent progress in hydrolase-catalyzed enantioselective transformations of carboxylic acid derivatives, alcohols, and epoxides. In light of the large number of such reactions in the literature, this review cannot be fully comprehensive; instead, the aim is to highlight the range of biotransformations that are possible. Emphasis has been placed on methodological advancements and recent applications to the synthesis of natural products and biologically active compounds. This chapter concentrates mainly on advances made during the period 2000 to mid2006. Unless indicated, only examples where the enantiomeric excess (ee) of the product is higher than 90% are included. This introduction brieﬂy presents general concepts that are essential for an understanding of the subject. For more complete coverage of these concepts and earlier literature, the reader is referred to a series of monographs [1–7]. 6.1.2 Reaction Conditions

Hydrolysis of substrates is performed in water, buffered aqueous solutions or biphasic mixtures of water and an organic solvent. Hydrolases tolerate low levels of polar organic solvents such as DMSO, DMF, and acetone in aqueous media. These cosolvents help to dissolve hydrophobic substrates. Although most hydrolases require soluble substrates, lipases display weak activity on soluble compounds in aqueous solutions. Their activity markedly increases when the substrate reaches the critical micellar concentration where it forms a second phase. This interfacial activation at the lipid–water interface has been explained by the presence of a

Asymmetric Organic Synthesis with Enzymes. Edited by Vicente Gotor, Ignacio Alfonso, and Eduardo Garcıa-Urdiales Copyright Ó 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 978-3-527-31825-4

j 6 Transesterification and Hydrolysis of Carboxylic Acid Derivatives, Alcohols, and Epoxides

134

helical oligopeptide (lid) that shields the active site. Upon interaction with a hydrophobic interface, the lid opens and the enzyme is rearranged into its active conformation [8]. Aqueous solutions are not suitable solvents for esteriﬁcations and transesteriﬁcations, and these reactions are carried out in organic solvents of low polarity [9–12]. However, enzymes are surrounded by a hydration shell or bound water that is required for the retention of structure and catalytic activity [13]. Polar hydrophilic solvents such as DMF, DMSO, acetone, and alcohols (log P < 0, where P is the partition coefﬁcient between octanol and water) are incompatible and lead to rapid denaturation. Common solvents for esteriﬁcations and transesteriﬁcations include alkanes (hexane/log P ¼ 3.5), aromatics (toluene/2.5, benzene/2), haloalkanes (CHCl3/2, CH2Cl2/1.4), and ethers (diisopropyl ether/1.9, tert-butylmethyl ether/ 0.94, diethyl ether/0.85). Exceptionally stable enzymes such as Candida antarctica lipase B (CAL-B) have been used in more polar solvents (tetrahydrofuran/0.49, acetonitrile/0.33). Room-temperature ionic liquids [14–17] and supercritical ﬂuids [18] are also good media for a wide range of biotransformations. Hydrolase-catalyzed reactions are usually performed over a narrow temperature range near room temperature, but recent studies have shown that some reactions proceed at extreme temperatures as low as 60 C or as high as 120 C [19–21]. Microwave irradiation is an alternative to conventional heating [22]. Hydrolase activity can also be improved through the use of additives such as surfactants, inorganic salts, amines, cyclodextrines, and crown ethers [23]. For large-scale or industrial applications, enzymes are immobilized or conﬁned to membrane-restricted compartments [24–27]. Immobilization methods can be grouped into several categories such as noncovalent binding by adsorption on inert supports, covalent immobilization on a solid carrier, and enzyme cross-linking with glutaraldehyde (CLEC). 6.1.3 Kinetic Resolution, Dynamic Kinetic Resolution, and Desymmetrization

A kinetic resolution depends on the fact that the two enantiomers of a racemic substrate react at different rates with the enzyme. The process is outlined in Figure 6.1, assuming that the (S) substrate is the fast-reacting enantiomer (kS > kR) and krac ¼ 0. In ideal cases, only one enantiomer is consumed and the reaction ceases at 50% conversion. In most cases, both enantiomers are transformed and the enantiomeric composition of the product and the remaining starting material varies with the extent

(S)-Substrate

kS

(S)-Product

Fast k rac (R)-Substrate

k S >>kR kR Slow

(R)-Product

Figure 6.1 Kinetic resolution (krac ¼ 0) and dynamic kinetic resolution (krac kS).

6.1 Introduction and General Aspects

O (a)

OH R

Hydrolase

R'

(r ac)

O O

O

OH R

R'

C 8F17

+

Organic layer

O

(b)

O–

O R

R'

O

Basic aqueous layer O

O

CAL-B

C 8 F17

O

OH +

(rac )

(S) Fluorous phase

(R) Organic phase

Figure 6.2 Separation of products by (a) cyclic anhydrides as acyl donors and (b) fluorous phase technique.

of conversion. The efﬁcacy of a kinetic resolution is measured by the enantiomeric ratio E ¼ kS/kR [28–30]. Resolutions are useful for synthesis when E > 20. Despite its widespread application [31,32], the kinetic resolution has two major drawbacks: (i) the maximum theoretical yield is 50% owing to the consumption of only one enantiomer, (ii) the separation of the product and the remaining starting material may be laborious. The separation is usually carried out by chromatography, which is inefﬁcient on a large scale, and several alternative methods have been developed (Figure 6.2). For example, when a cyclic anhydride is the acyl donor in an esteriﬁcation reaction, the water-soluble monoester monoacid is separable by extraction with an aqueous alkaline solution [33,34]. Also, ﬂuorous phase separation techniques have been combined with enzymatic kinetic resolutions [35]. To overcome the 50% yield limitation, one of the enantiomers may, in some cases, be racemized and resubmitted to the resolution procedure. Dynamic kinetic resolution (DKR), a more efﬁcient procedure, combines a kinetic resolution with an in situ racemization of a conﬁgurationally labile substrate (Figure 6.1, where krac kS) [36–38]. DKR has three main requirements: (i) the two enantiomers of the starting material must react at very different rates (kS kR), (ii) the starting material enantiomers must be in equilibrium (krac kS), and (iii) the product must be inert to racemization. Until recently, DKR was restricted to speciﬁc cases involving racemization through exchange of acidic hydrogen (hydantoins [39], oxazolones [40], thioesters [41]), halogen exchange (a-haloesters [42]), and addition– elimination reactions (cyanohydrins [43], reversible Michael addition [44]). The majority of chiral compounds resolved by hydrolase-catalyzed reactions have been secondary alcohols. Recently, the combination of hydrolases and transition-metal catalysts for DKR of secondary alcohols has attracted considerable attention [45–48]. Rutheniumbased organometallic catalysts are commonly used for the racemization of chiral secondary alcohols via hydrogen transfer (oxidation–reduction) as seen in Figure 6.3. Enantioselective enzymatic desymmetrization is the transformation of a substrate that results in the loss of a symmetry element that precludes chirality (plane of

j135

j 6 Transesterification and Hydrolysis of Carboxylic Acid Derivatives, Alcohols, and Epoxides

136

Ph H

OH Catalyst

R

R'

Ph

O R

Catalyst

HO

H

R

R'

R'

Catalyst =

Ph

Ph Ph OC Ru Cl CO

Figure 6.3 Racemization of a secondary alcohol in the presence of a ruthenium hydrogen-transfer catalyst.

(a)

A

B

Hydrolase

A

B

(b)

BA

A

B

X

X

Prochiral (plane of symmetry)

B

X

C

Y D

meso, achiral (plane of symmetry)

A

(c)

X D

X Y Chiral

X

BA

A

Hydrolase

C

Chiral

A X

Y Hydrolase

X

X

A Achiral (center of symmetry)

A Chiral

Figure 6.4 Hydrolase-catalyzed desymmetrization of a prochiral (a), a meso (b), or a centrosymmetric (c) substrate.

symmetry, center of symmetry) (Figure 6.4). In contrast to the classic kinetic resolutions of racemic mixtures, the theoretical maximum yield of these transformations is 100%. Enzymatic desymmetrization of prochiral or meso compounds to yield enantiomerically enriched products has proved to be a valuable tool in asymmetric synthesis [49,50]. The desymmetrizations of centrosymmetric substrates are very rare. In an asymmetric synthesis, the enantiomeric composition of the product remains constant as the reaction proceeds. In practice, however, many enzymatic desymmetrizations undergo a subsequent kinetic resolution as illustrated in Figure 6.5. For instance, hydrolysis of a prochiral diacetate ﬁrst gives the chiral monoalcohol monoester, but this product is also a substrate for the hydrolase, resulting in the production of R

R

(S)

(R)

OH

AcO Fast

R AcO

OAc (S)

Slow

Major

R Hydrolase

(R)

Achiral

R

R

(S)

(R)

HO

Slow

Fast

R

R

(S)

(R)

HO

OAc Minor

Figure 6.5 A desymmetrization coupled to a kinetic resolution.

OH Achiral

6.2 Enantioselective Biotransformations of Carboxylic Acid Derivatives

(S)-Substrate +

(S)-Product

(R)-Substrate Figure 6.6 Enantioconvergent transformation of a racemate.

the achiral diol. This second hydrolysis lowers the yield of the monoester but usually it increases its enantiomeric excess by kinetic resolution. The diacetate and the monoester are similar substrates and the enzymes usually have a preference for the same stereogenic center shown by arrows on Figure 6.5, thus removing the minor enantiomer. The second step is usually slower and the chiral product can be isolated in good yield if the reaction is monitored. When the chiral product is a monoacid monoester (HO2C-R-CO2R0 ), the process terminates after the ﬁrst step because polar carboxylic acids are hydrated in aqueous medium and are not substrates for hydrolases. 6.1.4 Enantioconvergent Transformation

An enantioconvergent transformation leads to a single enantiomeric product from a racemate [51]. Each enantiomer is transformed via independent pathways by the same catalyst or by two different catalysts (Figure 6.6). For example, the hydrolysis of epoxides may proceed with high regio- and stereoselectivity with inversion or retention of conﬁguration. Several enantioconvergent transformations of epoxides are reported in the last section of this chapter.

6.2 Enantioselective Biotransformations of Carboxylic Acid Derivatives 6.2.1 Ester Hydrolysis

Esterases, proteases, and some lipases are used in stereoselective hydrolysis of esters bearing a chiral or a prochiral acyl moiety. The substrates are racemic esters and prochiral or meso-diesters. Pig liver esterase (PLE) is the most useful enzyme for this type of reaction, especially for the desymmetrization of prochiral or meso substrates. The protected E-ring moiety of (S)-camptothecin has been prepared in enantiomerically enriched form by the enzymatic resolution of a triester with PLE in a pH 7 phosphate buffer-acetonitrile (5 : 1) solution (Figure 6.7). The alkaloid camptothecin is an inhibitor of the enzyme topoisomerase and some of its derivatives are anticancer drugs [52]. The resolution of racemic esters via selective hydrolysis catalyzed by hydrolases is a practical way to prepare optically active pharmaceutical intermediates as shown in Figure 6.8 [53,54].

j137

j 6 Transesterification and Hydrolysis of Carboxylic Acid Derivatives, Alcohols, and Epoxides

138

O

O

O

O

O

PLE O

EtO

CO 2Me OBn

N

O CO 2Me

EtO Phosphate buffer CH 3CN

N

OBn

HO

(R) + Mixture of acids

(r ac)

Camptothecin

O O

Figure 6.7 Chemoenzymatic synthesis of the E-ring portion of camptothecin.

O

H

CO2Me

CO2 Et

NH

(R) Asper gillus melleus protease buffer

S

t -Bu

O N

(R) K lebsiella oxyt oca hydrolase buf fer

(R) Bacillus brevis protease toluene buffer

CO 2(CH 2) 2OMe

Figure 6.8 Resolution of pharmaceutical intermediates (hydrolyzed enantiomer shown).

Several a-amino acids have been prepared from two-substituted malonates through an enzymatic desymmetrization Curtius rearrangement sequence (Figure 6.9). Hydrolysis of a triﬂuoroethyl cyclopropanedicarboxylate with PLE in pH 7 phosphate buffer provided the (R)-monoacid, which was rearranged to a cyclopropane a-amino acid via a Curtius-type reaction with diphenylphosphoryl azide [55]. The same strategy was applied to the synthesis of orthogonally protected (R)- and (S)-2methylcysteine. Desymmetrization by selective hydrolysis of a malonate with PLE in pH 7.2 phosphate buffer was followed by a Curtius rearrangement to selectively form the carbon–nitrogen bond. Interconversion of the ester and acid functionalities before the rearrangement allows the preparation of the other enantiomer [56]. The antiviral agent virantmycin is an unusual chlorinated tetrahydroquinoline isolated from a strain of Streptomyces (Figure 6.10). Hydrolysis of a prochiral 2,2disubstituted dimethyl malonate with PLE in DMSO-pH 8 phosphate buffer (1 : 4) was a key step in a stereodivergent synthesis of this natural product [57]. A [2 þ 2] photoaddition–cycloreversion was applied to the enantioselective synthesis of the natural product byssochlamic acid (Figure 6.11). Desymmetrization of a meso-cyclopentene dimethyl ester with PLE in pH 7 buffer-acetone (5 : 1) provided a monoacid, one of the photopartners. It is noteworthy that both enantiomers of this natural product were synthesized from the same monoacid [58]. CO 2CH2 CF3 CO 2CH2 CF3 t-BuS MeO2 C

PLE

CO 2H CO 2CH 2CF3

Buffer

PLE CO 2Me Buf fer

NH2

t-BuS

t-BuS HO2C

CO2 H

CO2 Me

H 2N

CO2 Me

Figure 6.9 Chemoenzymatic synthesis of amino acids via desymmetrization of malonates.

6.2 Enantioselective Biotransformations of Carboxylic Acid Derivatives

O MeO

OMe O

O

PLE

HO2C

DMSO buffer MeO

O

Cl

OH O

j139 O

N H

O (–)-Virantmycin

(S) Figure 6.10 Chemoenzymatic synthesis of virantmycin.

MeO 2C

PLE

O

O

HO2 C O

O

H 2O–acetone MeO2 C

MeO 2C

Byssochlamic acid

O O

Figure 6.11 Chemoenzymatic synthesis of byssochlamic acid.

A divergent synthesis of tropane alkaloid ferruginine was reported by Node and coworkers [59]. The b-ketoester intermediate was prepared by a novel PLE-catalyzed asymmetric dealkoxycarbonylation (hydrolysis followed by a decarboxylation) of a symmetric tropinone-type diester (Figure 6.12). Dimethyl sulfoxide was added to the phosphate buffer pH 8 (1 : 9) to reduce the activity of PLE and prevent over-dealkoxycarbonylation leading to tropinone. Several lipases were more efﬁcient than PLE and subtilisin Carlsberg for the desymmetrization of an N-t-butoxycarbonyl (Boc) meso-piperidine diester (Figure 6.13). The (3R)-monoester was converted into optically pure isogalactofagomine, a potent galactosidase inhibitor [60].

Bn

Bn

N

PLE

CO 2Et

Me

N

N

CO2 Et

DMSO buf fer EtO2 C O

O

O (+)-Ferruginine

Figure 6.12 Chemoenzymatic synthesis of (þ) or ()-ferruginine.

OH

OH CO2Me

MeO2C N Boc

Lipases H 2O

MeO 2C

OH CO 2H

3

OH

HO

5

N Boc (3R)

Figure 6.13 Chemoenzymatic synthesis of isogalactofagomine.

N H Isogalactof agomine

j 6 Transesterification and Hydrolysis of Carboxylic Acid Derivatives, Alcohols, and Epoxides

140

i-Bu

O

CAL-B O

i-Bu

i-Bu

O O n-Hexyl

n-Hexanol toluene

(r ac)

O

+ O

(R)

(S)

Figure 6.14 Alcoholysis of ibuprofen vinyl ester.

6.2.2 Ester Alcoholysis

Ester alcoholysis (transesteriﬁcation) in organic media is an equilibrium reaction and must be shifted in the desired direction. For example, Bornscheuer and coworkers [61] reported the resolution of ibuprofen vinyl ester by transesteriﬁcation with n-hexanol in the presence of CAL-B. The vinyl alcohol generated during the reaction tautomerizes to acetaldehyde, thus making the reaction irreversible, as illustrated in Figure 6.14. In some cases, the use of a large excess of alcohol is an option to drive the reaction to completion. Alcoholysis of glutamic acid dimethyl ester derivatives with acylase I was regio- and enantioselective (Figure 6.15). An excess of butanol was used as nucleophile and solvent [62]. Lipase-catalyzed methanolysis of racemic N-benzyloxycarbonyl (Cbz) amino acid triﬂuoroethyl esters carrying aliphatic side chains afforded the L-methyl esters and the D-triﬂuoromethyl esters (Figure 6.16). The released alcohol (CF3CH2OH) is a weak nucleophile that cannot attack the ester product. The nucleophilicity of the leaving group is depleted by the presence of an electron-withdrawing group [63]. 6.2.3 Esterification of Carboxylic Acids

The direct biocatalytic esteriﬁcation of a chiral acid with a simple achiral alcohol in organic media is a reversible process and, in order to bias the equilibrium to the

CO 2Me Acylase I R

CO2 Me (r ac)

BuOH

CO2 Me

CO2 Me +

R

R = PrCONH, CbzNH

CO 2Me (R)

R

CO 2Bu (S)

Figure 6.15 Alcoholysis of glutamic acid derivatives.

R

CO 2CH2CF 3 MeOH (4 equiv) R NHCbz (DL)

Carica papaya lipase

CO 2 CH3 NHCbz (L)

R

CO 2CH2CF 3

+

+ CF 3CH2 OH

NHCbz (D)

Figure 6.16 Enantioselective transesterification of N-Cbz-amino acid 2,2,2-trifluoroethyl esters.

6.2 Enantioselective Biotransformations of Carboxylic Acid Derivatives

O

O Burk holderia cepacia OH lipase

O OC 4H 9

1-Butanol hexane Na 2SO4

O (r ac)

j141 OH

+

O (S)

O (R)

Figure 6.17 Biocatalytic esterification in the presence of sodium sulfate as a drying agent.

product side, one of the reactant (alcohol) must be used in excess or one of the products (water) must be removed constantly during the reaction. (R)-3-Phenoxybutanoic acid and the corresponding butyl (S)-ester were obtained by Burkholderia cepacia lipase–catalyzed enantioselective esteriﬁcation of the racemic acid with 1-butanol in hexane containing anhydrous sodium sulfate to remove the water produced during the reaction (Figure 6.17) [64]. Orthoformates have been used in the lipase-catalyzed esteriﬁcation aimed at the kinetic resolution of racemic acids such as ﬂurbiprofen, a nonsteroidal anti-inﬂammatory drug (Figure 6.18). Orthoformates trap the water as it is formed through hydrolysis, and therefore prevent the reverse reaction, and, at the same time, provide the alcohol for the esteriﬁcation [65]. Several racemic methyl decanoic acids were resolved by esteriﬁcation with 1-hexadecanol in cyclohexane using immobilized Candida rugosa lipase (CRL) as the catalyst (Figure 6.19). The enantioselectivity was high even when the methyl group was as remotely located as in 8-methyldecanoic acid. Furthermore, for the substrates bearing the methyl group on an even-numbered carbon (i.e. for 2-, 4-, 6-, and 8-methyl decanoic acids), the enzyme showed (S) enantioselectivity and for

CO2H

+

MeCN, 45°C HC(OPr) 3

F (rac)

HC(OPr) 3 + H 2 O

CO2 H

CO2Pr

CAL-B F

F

HCO2 Pr + 2PrOH

Figure 6.18 Irreversible esterification in the presence of an orthoformate.

CO2H n

m

(r ac) Methyldecanoic acid

CRL C 16 H 33 OH Cyclohexane

(S)-2-,-4-,-6-,-8-Methyldecanoic ester + (R)-2-,-4-,-6-,-8-Methyldecanoic acid (R)-3-,-5-,-7-Methyldecanoic ester + (S)-3-,-5-,-7-Methyldecanoic acid

Figure 6.19 Resolution of 2- to 8-methyldecanoic acids.

j 6 Transesterification and Hydrolysis of Carboxylic Acid Derivatives, Alcohols, and Epoxides

142

OH

PSL O

O (r ac)

R

H2 O, buff er

O +

R

OH

O

(R)

R

O (S)

R = C5 H11, C6 H 13 Figure 6.20 Enantioselective hydrolysis of lactones.

odd-methyldecanoic acids (i.e. for 3-, 5-, and 7-methyl decanoic acids), the selectivity was reversed for the (R) substrate [66]. 6.2.4 Lactones

Enzyme-catalyzed stereoselective hydrolysis allows the preparation of enantiomerically enriched lactones. For instance, Pseudomonas sp. lipase (PSL) was found to be a suitable catalyst for the resolution of d-undecalactone and d-dodecalactone (Figure 6.20). Relactonization of the hydroxy acid represents an efﬁcient method for the preparation of both enantiomers of a lactone [67]. Asymmetric alcoholyses catalyzed by lipases have been employed for the resolution of lactones with high enantioselectivity. The racemic b-lactone (oxetan-2-one) illustrated in Figure 6.21 was resolved by a lipase-catalyzed alcoholysis to give the corresponding (2S,3S)-hydroxy benzyl ester and the remaining (3R,4R)-lactone [68]. Tropic acid lactone was resolved by a similar procedure [69]. These reactions are promoted by releasing the strain in the four-membered ring. Lipases also catalyze the intramolecular transesteriﬁcation (lactonization) of hydroxy esters. Macrolactonization of a racemic hydroxy ester in the presence of PSL provided the corresponding (R)-lactone (Figure 6.22). This compound is the naturally occurring enantiomer of the pheromone produced by the merchant grain beetle [70]. Chemical macrolactonizations require high dilution to minimize

C 3H 7

Lipase PS O

O (r ac)– Tr ans

C 3H 7 +

O

BnOH

C 3 H7

OBn OH

O (3R,4R)

O

(2S,3S)

Figure 6.21 Enzymatic alcoholysis of a b-lactone.

OH PSL 5

(rac)

CO2 Me

O Isooctane 60°C

Figure 6.22 Biocatalytic macrolactonization.

O (R)-Pheromone

+ (S)-Hydroxy ester

6.2 Enantioselective Biotransformations of Carboxylic Acid Derivatives

OH PPL BnO

OBn Et2O O

CO2Bn O

O

O

H

O

O

(S)-γ -Jasmolactone

Figure 6.23 Desymmetrization via a lipase-catalyzed lactonization.

Ph

Ph

H N

H N

O O

Ph

Ph

O O

O

CAL-B Ph

CH 3OH CH 3CN

Ph

N CO2 CH 3 H 88% (98% ee)

Figure 6.24 DKR of 2-phenyl-4-benzyl-5(4H)-oxazolone.

the competing oligomerizations. Biocatalytic macrolactonizations are temperaturesensitive and at lower temperatures (<45 C), oligomeric esters are the dominant products but better yields of lactones are obtained when the temperature is raised to 60–70 C, which indicate the higher energy of activation for lactonization. The ﬁrst asymmetric synthesis of ()-g-jasmolactone, a fruit ﬂavor constituent, was achieved via the enantioselective lactonization (desymmetrization) of a prochiral hydroxy diester promoted by porcine pancreas lipase (PPL) (Figure 6.23) [71]. Oxazolones (azlactones) are a form of activated lactones, so they are included in this section. CAL-B is an effective catalyst for the DKR of various racemic foursubstituted-5(4H)-oxazolones, in the presence of an alcohol, yielding optically active N-benzoyl amino acid esters as illustrated in Figure 6.24 [40]. Enantioselective biotransformations of lactides [72,73] and thiolactones [74] have also been reported. 6.2.5 Anhydrides

Enantioselective alcoholysis of racemic, prochiral, or meso cyclic anhydrides can be catalyzed by hydrolases, yielding the corresponding monoesters (Figure 6.25). In most cases, the enantioselectivity was moderate [75–77]. Organometallic catalysts or organocatalysts such as cinchona alkaloids are often more efﬁcient than enzymes for the stereoselective ring opening of cyclic anhydrides.

R2 R1

O

R1 Lipase R3 OH

R2

O O

O

Solvent R3 O

Figure 6.25 Enantioselective ring opening of anhydrides.

O *

OH

j143

j 6 Transesterification and Hydrolysis of Carboxylic Acid Derivatives, Alcohols, and Epoxides

144

Nitrile hydratase (NHase) R C N H2O

O R C NH 2

O R C OH

Amidase

+ NH 3

H2O

Nitrilase 2H2O Figure 6.26 Two pathways for biocatalytic hydrolysis of nitriles.

6.2.6 Hydrolysis of Nitriles

The hydrolysis of nitriles to carboxylic acids, which proceeds by way of the intermediate amide, requires harsh conditions such as strong bases or acids at a high temperature. Under appropriate conditions, the intermediate amide may be isolated as the major product. The transformation of nitriles by biocatalytic methods offers many advantages, including mild reaction conditions, chemoselectivity, regioselectivity, stereoselectivity, and the absence of by-products [78–81]. The biocatalytic hydrolysis of nitriles may follow two different pathways as illustrated in Figure 6.26. Nitrile hydratases (NHases, EC 4.2.1.84, classiﬁed in the lyase group) catalyzes the hydration of a nitrile to the corresponding amide, which may be converted by amidases (EC 3.5.1.4) to the carboxylic acid and ammonia. Nitrilases (EC 3.5.5.1) transform nitriles directly into acids. The lack of commercially available puriﬁed enzymes has prompted the use of crude enzyme preparations or whole microbial cells. One particularly interesting application of nitrile biohydrolysis is the largescale industrial production of acrylamide from acrylonitrile [82]. In the nitrile hydratase/amidase system, the enantiodiscrimination predominantly occurs during the hydrolysis of the intermediate amide by the amidase. Nitrilases and some nitrile hydratases can also be enantioselective. A wide range of a-alkyl arylacetonitriles were resolved by nitrile hydrolyzing enzymes [78]. 2-Arylpropionitriles were hydrolyzed to the corresponding arylpropionic acids related to nonsteroidal anti-inﬂammatory drugs such as ibuprofen. The biotransformation of 2-aryl-4pentenenitriles by Rhodococcus sp. microbial cells provided enantiopure 2-aryl-4pentenoic acids and their amide derivatives [83]. The potential applications of this reaction have been demonstrated by the preparation of (R)-()-baclofen, a speciﬁc g-aminobutyric acid receptor agonist and a potent antispastic agent (Figure 6.27). Enantioselective transformations of several cyclopropane or oxirane-containing nitriles were studied using nitrile-transforming enzymes [78]. Microbial Rhodococcus sp. whole cells containing a nitrile hydratase/amidase system hydrolyzed a number CN Rhodococcus sp.

Buffer

Cl ( rac )

H2 N

CONH 2 (S)-Acid

+ Cl

Cl (R )

Figure 6.27 Chemoenzymatic synthesis of (R)-()-baclofen.

CO2 H (R )-Baclofen

6.2 Enantioselective Biotransformations of Carboxylic Acid Derivatives

O Ar

O

Rhodococcus sp. CN

Buffer

(r ac)-tr ans

O CONH2 +

Ar

(2R,3S)

Ar

Ar

CO 2

(2S,3R)

of racemic trans-2,3-epoxy-3-arylpropionitriles to afford (2R,3S)-2-arylglycinamides (Figure 6.28). The arylglycidic acids underwent spontaneous decomposition to arylacetaldehydes under reaction conditions. The overall enantioselectivity originated from the combined effects of a dominantly high (2S)-selective amidase and low (2S)-selective nitrile hydratase [84]. Both cis- and trans-chrysanthemic nitriles and amides were resolved into highly enantiopure amides and acids by Rhodococcus sp. whole cells [85]. The overall enantioselectivity of reactions of nitriles originated from the combined effects of a higher (1R)-selective amidase and a (1R)-selective nitrile hydratase (Figure 6.29). Chrysanthemic acids are related to constituents of pyrethrum ﬂowers and insecticides. The addition of HCN to aldehydes or ketones produces cyanohydrins (a-hydroxy nitriles). Cyanohydrins racemize under basic conditions through reversible loss of HCN as illustrated in Figure 6.30. Enantiopure a-hydroxy acids can be obtained via the DKR of racemic cyanohydrins in the presence of an enantioselective nitriletransforming enzyme [86–88]. Many nitrile hydratases are metalloenzymes sensitive to cyanide and a nitrilase is usually used in this biotransformation. The DKR of mandelonitrile has been extended to an industrial process for the manufacture of (R)mandelic acid [89]. The synthesis of a-amino acids by reaction of aldehydes or ketones with ammonia and hydrogen cyanide followed by hydrolysis of the resulting a-aminonitrile is called the Strecker synthesis. Enzymatic hydrolysis has been applied to the kinetic resolution of intermediate a-aminonitriles [90,91]. The hydrolysis of (rac)-phenylglycine nitrile

R'

Rhodococcus sp.

CONH2

R

+

HO 2C

R

Buffer R

(r ac)–tr ans

R = CH 3 or Cl R' = CN, CONH 2

(1S)

R

(1R)

Figure 6.29 Biotransformation of chrysanthemic nitriles and amides.

OH R

CN (S)

O R

H + HCN

OH R

CN (R)

OH

Nitrilase H2 O

R

Figure 6.30 Nitrilase-catalyzed dynamic kinetic resolution of cyanohydrins.

CHO +

CO 2H

Figure 6.28 Enantioselective hydrolysis of racemic trans-2,3-epoxy-3-arylpropanenitriles.

R

j145

CO2 H

R

j 6 Transesterification and Hydrolysis of Carboxylic Acid Derivatives, Alcohols, and Epoxides

146

NH2

NH2

Rhodococcus sp.

Ph CN (rac)

Nitrile hydratase

Ph C (r ac) O

NH 2

Amidase

NH 2

Ph

C

NH 2 NH2 +

Ph

C

OH

(L) O

(D) O

Figure 6.31 Kinetic resolution of phenylglycine nitrile.

OH NC

CN

Nitrilase

OH NC

OH

EtOH CO 2H

(R)

H+

NC

CO2 Et

Figure 6.32 Desymmetrization of a meso-dinitrile by a nitrilase.

catalyzed by a Rhodococcus strain containing a nonstereoselective nitrile hydratase and a highly selective amidase gave enantiomerically pure D-phenylglycine amide and L-phenylglycine [91]. This example is illustrated in Figure 6.31. The biocatalytic differentiation of enantiotopic nitrile groups in prochiral or meso substrates has been studied by several research groups. For instance, the nitrilasecatalyzed desymmetrization of 3-hydroxyglutaronitrile [92,93] followed by an esteriﬁcation provided ethyl-(R)-4-cyano-3-hydroxybutyrate, a useful intermediate in the synthesis of cholesterol-lowering drug statins (Figure 6.32) [94,95]. The hydrolysis of prochiral a,a-disubstituted malononitriles by a Rhodococcus strain expressing nitrile hydratase/amidase activity resulted in the formation of (R)-a,a-disubstituted malonamic acids (Figure 6.33) [96]. 6.2.7 Hydrolysis of Amides, Lactams, and Hydantoins

The main application of the enzymatic hydrolysis of the amide bond is the enantioselective synthesis of amino acids [4,97]. Acylases (EC 3.5.1.n) catalyze the hydrolysis of the N-acyl groups of a broad range of amino acid derivatives. They accept several acyl groups (acetyl, chloroacetyl, formyl, and carbamoyl) but they require a free a-carboxyl group. In general, acylases are selective for L-amino acids, but Dselective acylase have been reported. The kinetic resolution of amino acids by acylasecatalyzed hydrolysis is a well-established process [4]. The in situ racemization of the substrate in the presence of a racemase converts the process into a DKR. Alternatively, the remaining enantiomer of the N-acyl amino acid can be isolated and racemized via the formation of an oxazolone, as shown in Figure 6.34. A recent example of the acylase method is shown in Figure 6.35. N-Benzyloxycarbonylproline is resolved by a proline acylase in a strain of Arthrobacter sp. The

n

Ar NC (r ac)

Rhodococcus sp. CN Nitrile hydratase n = 1, 2

n

Ar H 2 NOC

Amidase CONH2

(r ac)

Figure 6.33 Desymmetrization of prochiral malononitriles.

n

Ar HO2C (R)

CONH2

6.2 Enantioselective Biotransformations of Carboxylic Acid Derivatives

Racemase (in situ)

COOH

COOH

Acylase

NHAc

H2 N

Buffer

R (r ac )

COOH NHAc R ( D)

+ R

( L)

Ac2 O R

O

R

H 2O

O

Base N

O

N

O

Figure 6.34 Resolution of N-acyl amino acids by acylases.

CO2 H

N O

(DL)

O

Acylase

CO2 H

N

Ar throbact er sp. O

O

(D)

Ph

N H

+

CO 2H

(L)

+ HO

Ph

Ph

+

CO2

Figure 6.35 Acylase-catalyzed resolution of Cbz-(D,L)-proline.

carbamic acid formed during the reaction spontaneously decomposes into CO2 and benzyl alcohol [98,99]. Penicillin G acylase (PGA, EC 3.5.1.11, penicillin G amidase) catalyzes the hydrolysis of the phenylacetyl side chain of penicillin to give 6-aminopenicillanic acid. PGA accepts only phenylacetyl and structurally similar groups (phenoxyacetyl, 4-pyridylacetyl) in the acyl moiety of the substrates, whereas a wide range of structures are tolerated in the amine part [100]. A representative selection of amide substrates, which have been hydrolyzed in a highly selective fashion, is depicted in Figure 6.36. The natural role of peptide deformylases (EC 3.5.1.31) is the hydrolysis of the N-terminal formyl group from nascent peptides. This enzyme combines a good

O O

NH O

HO

R

O Ph

HN

O R'

Ph (S, S) R = Me, Et, CH 2CH=CH2 , Bn

(R) R' = CH 2OH, COOH

Figure 6.36 Examples of amides resolved by penicillin G acylase– catalyzed hydrolysis (the fast-reacting enantiomer is shown).

Ph

OEt N H

(R) R'' = H, TMS

R''

j147

j 6 Transesterification and Hydrolysis of Carboxylic Acid Derivatives, Alcohols, and Epoxides

148

R

CO2 H

HN (r ac)

H

R

Peptide deformylase

CO2 H

HN

H 2O

H

(D )

O

R

CO 2H

+

NH2 (L)

O

Figure 6.37 Peptide deformylase-catalyzed hydrolysis of N-formyl amino acids.

activity with a very high enantioselectivity in the resolution of N-formylated a- and b-amino acids, a-amino acid amides and a-aminonitriles (Figure 6.37) [101]. Racemic a-amino amides and a-hydroxy amides have been hydrolyzed enantioselectively by amidases. Both L-selective and D-selective amidases are known. For example, a puriﬁed L-selective amidase from Ochrobactrum anthropi combines a very broad substrate speciﬁcity with a high enantioselectivity on a-hydrogen and a,a-disubstituted a-amino acid amides, a-hydroxyacid amides, and a-N-hydroxyamino acid amides [102]. A racemase (a-amino-e-caprolactam racemase, EC 5.1.1.15) converts the D-aminopeptidase-catalyzed hydrolysis of a-amino acid amides into a DKR (Figure 6.38) [103]. The hydrolysis with a (R)-speciﬁc amidase from Klebsiella oxytoca resolved 2-hydroxy- and 2-amino-3,3,3-triﬂuoro-2-methylpropionamide, giving the (R)-acid and the remaining (S)-amide (Figure 6.39) [104,105]. b-Lactamases (EC 3.5.2.6) produced by bacteria cleave the b-lactam ring and are responsible for their resistance to b-lactam antibiotics. Lactamases are useful catalysts for the enantioselective hydrolysis of b-lactams and other cyclic amides. b-lactams shown in Figure 6.40 were resolved by whole-cell systems containing an amidase [106]. 2-Azabicyclo[2.2.1]hept-5-en-3-one, a bicyclic g-lactam, is an intermediate for the synthesis of antiviral carbocyclic nucleosides (Figure 6.41). This compound was resolved using a cloned lactamase in an industrial-scale process [106,107]. Lipases are generally inactive on the amide bond. One notable exception is the hydrolysis and alcoholysis of b-lactams by CAL-B [108–112]. For example, CAL-B O

NH 2

H2 N

O

NH 2

Racemase

H

H

R (L )

O

NH 2

H

R (D )

X

Amidase NH 2 O

(r ac)

H 2O X = OH, NH2

NH2 R (D )

Figure 6.38 Dynamic kinetic resolution of amino acid amides.

F 3C

OH

D -Aminopeptidase

X

X HO

CF3

+

F3C

NH 2

O

O

(R)

(S)

Figure 6.39 Resolution of 2-hydroxy- and 2-amino-3,3,3-trifluoro2-methylpropionamide by an amidase.

6.2 Enantioselective Biotransformations of Carboxylic Acid Derivatives R

S

NH

S

O

R

O

j149

S

NH

O

R

N H

Figure 6.40 b-Lactams resolved by lactamase-catalyzed hydrolysis (the reactive enantiomer is shown).

N H (rac)

O Lactamase

HN

NH

NH2 +

HO2C

N N

O

(+)

N

(–)

HO

N

NH 2

Abacavir

Figure 6.41 A bicyclic g-lactam resolved by a lactamase.

CO2 H

CAL-B O N H (r ac)

NH 2

H 2O/i-Pr2O 60°C

O

+

HN

(1R,2S)-Cispentacin

(1S,5R)

Figure 6.42 Lipase-catalyzed kinetic resolution of a b-lactam.

catalyzed the enantioselective ring opening of several b-lactams in water-saturated organic media at 60 C as exempliﬁed in Figure 6.42 [110]. Racemic hydantoins result from the reaction of carbonyl compounds with potassium cyanide and ammonium carbonate or the reaction of the corresponding cyanohydrins with ammonium carbonate (Bucherer–Bergs reaction). Hydantoins racemize readily under basic conditions or in the presence of hydantoin racemase, thus allowing DKR (Figure 6.43). Hydantoinases (EC 3.5.2.2), either isolated enzymes or whole microorganisms, catalyze the hydrolysis of ﬁve-substituted O

H N

Spontaneous racemization (OH – )

O

H N

O R

N H ( L)

O or hydantoin racemase

R

N H (D )

HO

O

R

N H

D-selective

Hydantoinase

O

N-Carbamoyl-L-Amino

acid CO 2H NH 3 + CO 2 +

NH2

R NH 2 D-Amino acid

HNO2/HCl or carbamoylase

Figure 6.43 Dynamic kinetic resolution of (rac)-hydantoins by a D-hydantoinase.

j 6 Transesterification and Hydrolysis of Carboxylic Acid Derivatives, Alcohols, and Epoxides

150

hydantoins to N-carbamoyl a-amino acids [39,113–115]. Most hydantoinases are D-enantioselective and convert hydantoins into nonnatural D-amino acid derivatives. The N-carbamoyl group is removed chemically (HNO2/HCl) or biochemically with a carbamoylase (EC 3.5.1.d, carbamoyl amino acid amidohydrolase). D-Phenylglycine and D-4-hydroxyphenyl glycine for the semisynthetic penicillins are produced commercially by the hydantoinase method.

6.3 Enantioselective Enzymatic Transformations of Alcohols

The principal methods for the hydrolase-promoted synthesis of enantiomerically pure alcohols are depicted in Figure 6.44. Biocatalytic acylation and alcoholysis have been reviewed recently [116,117]. Lipases, esterases, and proteases catalyze these reactions, but CAL-B [118–120], CRL [121,122], and diverse lipase preparations from Pseudomonas species are common place. In hydrolytic reactions, water (solvent and nucleophile) is in large excess and the equilibrium is shifted toward the products. In contrast, alcoholysis of esters and acylation of alcohols are reversible transesteriﬁcations. The use of an excess of the nucleophile (alcohol) or the removal of the ester product are options to drive alcoholysis to completion. For acylation reactions, the use of activated esters shifts the equilibrium in favor of the products [123,124]. In the case of quasi-irreversible acyl donors such as 2-haloethyl, cyanomethyl, and oxime esters, the released alcohol is a weak nucleophile that cannot attack the ester product. The nucleophilicity of the leaving group is depleted by the presence of electron-withdrawing groups. The most common activated acyl donors are enol esters such as vinyl acetate (VA), isopropenyl acetate (IPA), and 1-ethoxyvinyl esters. The product alcohol is an enol that tautomerizes to a carbonyl compound (acetaldehyde or acetone) or ethyl acetate. VA is more reactive but the liberated acetaldehyde slowly inactivates some lipases (e.g. CRL), probably by formation of imines with side chains of lysines. 1-Ethoxyvinyl esters are as reactive as vinyl esters and the leaving group, ethyl acetate, is inert toward enzymes [125]. Anhydrides (linear, cyclic, mixed carboxylic-carbonic) are also useful acyl donors for irreversible acylations (Figure 6.45). O R*

O

Hydrolysis

+

H 2O

+

R''-OH

R O

R*

O

+

R-CO2 H

R*-OH

+

R''

O

Alcoholysis R

Organic solvent O

R*-OH

R*-OH

+ R'

O

R* Organic solvent

R

O

Acylation R

O

+ O

Figure 6.44 Types of enzyme-catalyzed reactions leading to enantiopure alcohols. Stereogenic centers are included in the alcohol moiety, as indicated by an asterisk.).

R

R'-OH

6.3 Enantioselective Enzymatic Transformations of Alcohols

O R*-OH

+

R

R'

O O

R*

R'

+ R

O

O

HO

R'

R' = H, CH 3 , O Alkyl O R*-OH

+

R

O

O O

R* R'

+ O

R

R'-CO2 H

Figure 6.45 Activated acyl donors for irreversible acylation of alcohols.

Acyl migrations prevailing under aqueous or protic conditions are slower in aprotic organic media. Accordingly, acylation usually gives better results than hydrolysis. Aromatic acyl donors are used when the product is sensitive to acyl migration. Aromatic acyl groups migrate less than aliphatic ones, since the formation of the tetrahedral intermediate involves a greater loss in resonance energy. When both the alcohol and the corresponding ester are substrates for an enzyme, acylation and deacylation (hydrolysis or alcoholysis) reactions are usually complementary and give the opposite enantiomers. Although hydrolysis/alcoholysis and acylation represent reactions in opposite directions, the enzyme favors the same enantiomer (or the same prochiral group) in both cases. This empirical rule applies to kinetic resolutions and desymmetrizations but a few exceptions have been reported [126]. For example, the acylation of meso-tetrahydropyrans (Figure 6.46) by VA in the presence of CAL-B in organic media and the hydrolysis of the corresponding diacetates catalyzed by the same enzyme provided the opposite enantiomers of the monoesters [127]. Both enantiomers of isolevoglucosenone have been prepared by complementary hydrolysis and acylation reactions (Figure 6.47). ()-Levoglucosenone,

OR

OR

OR

OR

CAL-B

OH

CAL-B R

VA

O OH

R

S

S

OAc

Buffer

O

O OH

OH

OAc

Acylation

O OAc

OAc

Hydrolysis

Figure 6.46 Desymmetrization: acylation and hydrolysis yield opposite enantiomers.

O

OH

Lipase AK

O

VA

O

OAc O

(r ac)

O

(R,R,R) Lipase AK Buffer acetone

MnO2

O

+

O

(rac) O

OAc

O

+ O OAc

(S,S,S)

O O

OH

(S,S,S) O

j151

O

(S,S) OH

MnO 2

O

O

(R,R,R)

Figure 6.47 Kinetic resolution: acylation and hydrolysis yield opposite enantiomers.

O

(R,R)

O

j 6 Transesterification and Hydrolysis of Carboxylic Acid Derivatives, Alcohols, and Epoxides

152

OH Bu 3Sn

OH Ph

Lipozyme VA, i-Pr2 O

OH

Se

Te

Ph

CAL-B VA, hexane

CAL-B VA, hexane

Figure 6.48 Favored enantiomer in lipase-catalyzed acylations of racemic alcohols containing an organometallic substituent.

a regioisomer of isolevoglucosenone, is obtained in low yield by an acid-catalyzed pyrolysis of cellulose. Isolevoglucosenone can be converted into levoglucosenone and both compounds are versatile chiral building blocks [128]. The wide substrate tolerance of lipases is demonstrated by the resolution of organometallic substrates [129–131]. The presence of tin, selenium, or tellurium in the structure of secondary alcohols does not inhibit the lipase activity and enantiopure organometallic alcohols were obtained by acylation in organic media (Figure 6.48). Sulfatases catalyze the hydrolysis of sulfate esters as shown in Figure 6.49. Arylsulfatases generally act through retention of conﬁguration by cleavage of the S–O bond, whereas some alkyl sulfatases break the C–O bond of sulfate esters leading to inversion of conﬁguration. The enantioselective hydrolysis of sec-alkyl sulfate esters by alkyl sulfatases contained in whole cells of Sulfolobus species proceeds with inversion of conﬁguration and furnishes a homochiral product mixture [132]. Biocatalysis has emerged as an important tool for the enantioselective synthesis of chiral pharmaceutical intermediates and several review articles have been published in recent years [133–137]. For example, quinuclidinol is a common pharmacophore of neuromodulators acting on muscarinic receptors (Figure 6.50). (R)-Quinuclidin-3ol was prepared via Aspergillus melleus protease-mediated enantioselective hydrolysis of the racemic butyrate [54,138]. Calcium hydroxide served as a scavenger of butyric acid to prevent enzyme inhibition and the unwanted (R) enantiomer was racemized over Raney Co under hydrogen for recycling.

OSO 3 –Na+ R

+

Sulf olobus sp. whole cells

R'

OSO 3– Na+ R

OH

(Sulfatase) buff er

R'

R

+

R' -

+

OSO3 Na R

R = CH 3, R' = n-C 6 H13 H 3O+ R = CH , R' = CH -Ph 3 2

R'

Figure 6.49 Deracemization of sec-alkyl sulfate esters.

Asper gillus melleus protease

n-PrCO 2 N

buffer

HO

n-PrCO2 + N (S)

Figure 6.50 Protease-catalyzed resolution of quinuclidin-3-ol.

N (R)

6.3 Enantioselective Enzymatic Transformations of Alcohols

HO

Lipase, VA AcO

NHCbz

CH2 Cl2

(r ac)

O

+ (–)-Alcohol

CO2 H HN

Lipase AK

OH

NHCbz

O

OAc

(–)-Kainic acid

VA

(rac)

CO2 H

Figure 6.51 Chemoenzymatic synthesis of kainic acid.

Ogasawara and coworkers reported a concise route to ()-kainic acid (Figure 6.51), an excitatory neurotransmitter of marine origin, via a lipase-mediated kinetic resolution of an N-Cbz aminocyclopentenol [139]. More recently, ()-kainic acid was synthesized from an optically active butenolide prepared by enzymatic DKR [140]. Conduritols and inositols are cyclic polyalcohols with signiﬁcant biological activity. The presence of four stereogenic centers in the structure of conduritols allows the existence of 10 stereoisomers. Enzymatic methods have been reported for the resolution of racemic mixtures or the desymmetrization of meso-conduritols. For example, Mucor miehei lipase (MML) showed enantiomeric discrimination between all-(R) and all-(S) stereoisomers of conduritol E tetraacetate (Figure 6.52). Alcoholysis resulted in the removal of the four acetyl groups of the all-(R) enantiomer whereas the all-(S) enantiomer was recovered [141]. Gabosines are secondary metabolites isolated from various Streptomyces strains. A rational approach to the synthesis of gabosines and related carbosugars includes the enantioselective acylation of a hydroxy ketal (masked p-benzoquinone) (Figure 6.53). The strategy has been applied to the synthesis of both enantiomers of gabosine N and O and established the absolute conﬁguration of natural gabosine O [142].

OAc

OH

OAc OAc

OAc +

OAc

OAc

OAc

OAc

All-(R)

All-(S)

OAc OH

MML BuOH

OH

t-BuOMe OH All-(R)

OH CAL-B

O

O (r ac)

SPh

VA i-Pr2 O

OH

(R)

O

HO

O

SPh

+(S)-Acetate

Figure 6.53 Chemoenzymatic synthesis of gabosine O.

OAc OAc

Figure 6.52 Enzymatic alcoholysis of racemic conduritol E tetraacetate.

OH

OAc +

HO O (–)-Gabosine O

All-(S)

j153

j 6 Transesterification and Hydrolysis of Carboxylic Acid Derivatives, Alcohols, and Epoxides

154

O

+ O N (r ac) + O

–

EtO

OH

O

CAL-B CH 3CN

O

CO 2Et

H OH

N

(R)

CO2 Et

OH O

H

O

N+

N CO2 Et

OH

O-

O

(–)-Rosmarinecine

Figure 6.54 A domino DKR-intramolecular 1,3-dipolar cycloaddition reaction.

The simple acyl moiety introduced during the enzymatic acylation of chiral alcohol is usually removed or replaced by other more complex groups during subsequent reactions. Kita and coworkers [143,144] developed sequential transformations in which the acyl moiety bearing functional groups is utilized as part of the constituent structure in a subsequent intramolecular reaction. Following this concept, they reported a concise asymmetric total synthesis of ()-rosmarinecine [145]. The domino reaction was a DKR of a hydroxy nitrone followed by an intramolecular 1,3-dipolar cycloaddition (Figure 6.54). Combination of lipase-catalyzed transesteriﬁcation with unsaturated vinyl esters as acyl donors and ring-closing metatheses (RCMs) have also been reported [146–148]. Two groups applied this strategy for the synthesis of goniothalamin from cinnamaldehyde [147,148]. The key steps were a transesteriﬁcation using vinyl acrylate as acyl donor, followed by an RCM, as depicted in Figure 6.55. Enzymatic desymmetrization of prochiral or meso-alcohols to yield enantiopure building blocks is a powerful tool in the synthesis of natural products. For example, a synthesis of conagenin, an immunomodulator isolated from a Streptomyces, involved two enzymatic desymmetrizations [149]. The syn–syn triad of the acid moiety was prepared via a stereoselective acylation of a meso-diol, whereas the amine fragment was obtained by the PLE-catalyzed hydrolysis of a prochiral malonate (Figure 6.56).

O O

OH

O O

O RCM

CAL-B Hexane

(r ac)

O

(R) + (S)-Alcohol

Goniothalamin

Figure 6.55 Combination of lipase-catalyzed transesterification and RCM.

Cbz-HN

CO2 Et

PLE

Cbz-HN

CO2 Et Buff er, CH3 CN Lipase HO

OH

CO2 Et CO2 H

AcO

OH

VA, THF Figure 6.56 Convergent chemoenzymatic synthesis of (þ)-conagenin.

OH

OH

H N

CO2 H

O (+)-Conagenin

OH

j155

6.3 Enantioselective Enzymatic Transformations of Alcohols

OAc

OAc

PLE H 2O

AcO

(+)

HO

Figure 6.57 Desymmetrization of a centrosymmetric diester.

Few examples of chemical or enzymatic desymmetrizations of centrosymmetric molecules (point group S2 ¼ Ci) have been described [150]. The PLE-catalyzed hydrolysis of a centrosymmetric cyclohexanediacetate gave rise to an enantiomerically pure (99.5% ee) cyclohexanediol monoacetate in high yield [151] (Figure 6.57). Polypropionate chains with alternating methyl and hydroxy substituents are structural elements of many natural products with a broad spectrum of biological activities (e.g. antibiotic, antitumor). The anti–anti stereotriad is symmetric but is the most elusive one. Harada and Oku described the synthesis and the chemical desymmetrization of meso-polypropionates [152]. More recently, the problem of enantiotopic group differentiation was solved by enzymatic transesteriﬁcation. The synthesis of the acid moiety of the marine polypropionate dolabriferol (Figure 6.58a) and the elaboration of the C(19)–C(27) segment of the antibiotic rifamycin S (Figure 6.58b) involved desymmetrization of meso-polypropionates [153,154]. The desymmetrization principle was also exploited in the synthesis of (þ)FR900482, an antitumor antibiotic isolated from Streptomyces sandaensis [155]. A prochiral propanediol was enzymatically desymmetrized using PSL to give the corresponding (S)-monoester as illustrated in Figure 6.59. The stereoselective acylation of a prochiral chromanedimethanol derivative by VA in the presence of CAL-B gave the corresponding (S)-monoester in high enantiomeric

(a)

CRL HO

OH OTBS

HO

OAc

VA Hexane

OTBS

O

CO2 H OTBS

Lipases

(b) OH

OH

OTBS OH

VA

OH

OAc OH

OTBS OH

OH

Figure 6.58 Enzymatic desymmetrization of meso-polypropionates.

OBn

OH OH

OBn

PSL

OH

OH

OH

OAc

VA NO2 OBn

OCONH2

NO 2 OBn

(S)

OHC

O N (+)-FR900482

Figure 6.59 Application of enzymatic desymmetrization to the synthesis of (þ)-FR900482.

NH

j 6 Transesterification and Hydrolysis of Carboxylic Acid Derivatives, Alcohols, and Epoxides

156

HO

CAL-B OH

O

OH

HO O

VA Et2O

(R)- α-Tocotrienol

OAc OH

HO

HO O

OH

O (R)

(S)-α-Tocotrienol

Figure 6.60 Chemoenzymatic synthesis of a-tocotrienol.

O O

O CH 2OAc

Phosphate buffer DMSO (10%)

O O O

PPL

CH 2 OH

O O O

OH

CH 2OAc

O

CH 2 OAc

MeO

OMe OMe

(–)-Podophyllotoxin

Figure 6.61 Chemoenzymatic synthesis of ()-podophyllotoxin.

purity (Figure 6.60). In contrast, enzymatic hydrolysis of prochiral diesters failed to give the (R)-monoester in good yield and high enantiomeric excess. Thus, both enantiomers of a-tocotrienol, one of the main vitamin E isoforms, were synthesized from the (S)-monoester [156]. Podophyllotoxin, a plant lignan, is a potent antimitotic agent (Figure 6.61). An enantioselective synthesis of ()-podophyllotoxin was achieved via the enzymatic desymmetrization of an advanced meso-diacetate, through PPL-mediated diester hydrolysis [157]. Stereoselective enzymatic transformation of substrates generated from Diels– Alder adducts is a versatile method for the preparation of complex chiral building blocks. The 1,4-benzoquinone-cyclopentadiene adduct has been the starting material for the chemoenzymatic synthesis of a variety of natural products as illustrated in Figure 6.62. Desymmetrization of an epoxyquinonediol (Figure 6.62a) was a key step in the synthesis of several polyketide-derived natural products such as cycloepoxydon [158]. Reduction of the adduct and desymmetrization of the corresponding meso-ene1,4-diol (Figure 6.62b) provided a chiral intermediate for the synthesis of the aminocyclitol core of hygromycin A [159,160]. Ogasawara and coworkers optimized the ring contraction of the adduct epoxide (Figure 6.62c) and synthesized tanikolide, a marine antifungal compound, via a stereoselective acylation and a retro Diels–Alder reaction [161]. A similar methodology was applied to the enantioselective total synthesis of epoxyquinone natural products such as epoxydon [162] (Figure 6.62d).

6.4 Hydrolysis of Epoxides

O

O

(a) HO

Lipase PS

AcO

VA, THF 0 °C

HO

O

HO

j157

meso O

O O

O O

O OH

OH

Cycloepoxydon (b)

H

O

H

O

OAc

H

2. Enzymatic desymmetrization

O

H

H 2N

H

OH

Hygromycin A subunit

O

OH

H

H

O

meso

TBSO

HO

OAc

Lipase LIP

O H

OH

OH

p-Benzoquinonecyclopentadiene adduct

(c)

O

HO

1. Reduction

O

O C11H 23

VA, Et2 O OH

AcO +

(r ac)

(+)-Tanikolide

Primary (+)-monoacetate

O

H

OH

OAc

Lipase PS

(d)

O

O

O

VA H

OH

TBSO

O

(r ac)

OH

+ (–)-Alcohol

Figure 6.62 Chemoenzymatic synthesis of natural products from the p-benzoquinone-cyclopentadiene adduct.

6.4 Hydrolysis of Epoxides

Chiral epoxides and their corresponding vicinal diols are very important intermediates in asymmetric synthesis [163]. Chiral nonracemic epoxides can be obtained through asymmetric epoxidation using either chemical catalysts [164] or enzymes [165–167]. Biocatalytic epoxidations require sophisticated techniques and have thus far found limited application. An alternative approach is the asymmetric hydrolysis of racemic or meso-epoxides using transition-metal catalysts [168] or biocatalysts [169–174]. Epoxide hydrolases (EHs) (EC 3.3.2.3) catalyze the conversion of epoxides to their corresponding vicinal diols. EHs are cofactor-independent enzymes that are almost ubiquitous in nature. They are usually employed as whole cells or crude

O

Epoxydon

j 6 Transesterification and Hydrolysis of Carboxylic Acid Derivatives, Alcohols, and Epoxides

158

a b

O R

EH, H 2O

a Retention b

kS >> kR

Inversion

(r ac)

(R )

O

+

R

(S)

R (R)

O

HO HO

R

+

HO HO

(R)

R

Figure 6.63 Stereo- and regioselectivity of epoxide hydrolysis.

cell-free extracts, but a few (e.g. enzymes from Aspergillus niger and Rhodococcus rhodochrous) have recently become commercially available. Parallel to the discovery of new natural EHs [175,176], directed evolution techniques combined with highthroughput screening [177–179] can generate novel enzymes with improved properties. Epoxide biohydrolyses are usually performed in aqueous media; recent research has, however, demonstrated that A. niger EH is more stable and retains some activity in organic media [180]. The efﬁciency of EHs is sometimes limited by the low solubility of hydrophobic epoxides in aqueous media or the poor stability of the substrate resulting in spontaneous nonenzymatic hydrolysis. The structure and function of several different EHs have been investigated [169,181,182]. The simpliﬁed mechanism involves a trans antiperiplanar addition of water to the oxirane ring leading to the trans 1,2-diols. In general, the attack on a stereogenic center leads to an inversion of conﬁguration. Since the ring opening can occur at two different carbon atoms, the regioselectivity of the enzymatic attack is also an important factor. The stereochemical pathways of hydrolysis of monosubstituted epoxides are illustrated in Figure 6.63. The ring opening may proceed via an attack on the less substituted carbon (attack a) leading to retention of conﬁguration or at the stereogenic center (attack b) with inversion of conﬁguration. Thus the absolute conﬁgurations of both the diol and the remaining substrate have to be determined separately. In styrene oxide–type epoxides, the attack at the benzylic position is electronically facilitated by resonance stabilization of the carbocation in the transition state. Activity and selectivity vary with the enzyme source and the substrate structure (substitution pattern, steric crowding, and electronic factor) and one cannot make broad generalizations. Pyridyl oxiranes (Figure 6.64) were stereoselectively hydrolyzed by EH from A. niger to give the (S)-epoxide and the (R)-diol [183]. The reactions proceed with high regio- and enantioselectivity on the least hindered carbon atom of the

O

R H

(S) +

H R

O

Asper gillus niger EH

R H

O (S)

+ H R

OH OH

R=

N

(R) N

(R)

Figure 6.64 Kinetic resolution of pyridyl oxiranes.

N

j159

6.4 Hydrolysis of Epoxides

i-Bu

i-Bu

i-Bu i-Bu

A. niger

O

+ EH

O

OH

1.HBr/AcOH

(r ac)

(R)

OH

O

OH

(S)

(S)-Ibuprofen

2.KOH/MeOH Figure 6.65 Chemoenzymatic synthesis of (S)-ibuprofen.

(R)-epoxide. It is noteworthy that this stereoselective reaction cannot be accomplished with transition-metal catalysts. The hydrolysis of various para-substituted a-methylstyrene oxides was studied using 10 EHs [184]. The hydrolysis of the isobutyl compound with the enzyme from A. niger was the key step in the synthesis of (S)-ibuprofen (Figure 6.65). The (R)-diol was recycled via chemical racemization. Enzymatic resolution of epoxides, like other resolution methods, affords a maximum of 50% yield of optically pure product based on racemic starting material. This limitation can be overcome by enantioconvergent processes, in which each of the substrate enantiomers is converted into the same product enantiomer via two independent pathways. Enantioconvergence has been achieved using (i) two biocatalysts, (ii) a single biocatalyst, or (iii) a combination of a bio- and a chemocatalyst. Both A. niger and B. sulfurescens EHs catalyze the enantioselective hydrolysis of rac-styrene oxide [185] (Figure 6.66). These two microorganisms present opposite enantioselectivities toward the epoxide and the absolute conﬁguration of the product diol is R in both cases. Thus, the enantioconvergent biohydrolysis with a mixture of the two fungi in the same reactor provides the (R)-diol in high yield and high enantiomeric excess. Enantioconvergent hydrolysis of para-chlorostyrene oxide (Figure 6.67) using a one-pot sequential bienzymatic strategy provided the corresponding (R)-diol in high yield (96% ee, yield ¼ 93%) [186]. The second enzyme was added after about 50% conversion because the ﬁrst one was sensitive to inhibition by the (R)-diol.

A.niger

O a +

b O

a Retention B. sulf ur escens b Inversion

(r ac) Both microorganisms Single reactor

OH O (S) (23%, 96% ee)

OH (R) (54%, 51% ee) OH

O OH (R) (19%, 98% ee)

(R) (47%, 83% ee)

OH (R ) (92%, 89% ee) OH

Figure 6.66 Enantioconvergent hydrolysis of styrene oxides using two biocatalysts.

j 6 Transesterification and Hydrolysis of Carboxylic Acid Derivatives, Alcohols, and Epoxides

160

O

O

OH

HO

Solanum t uber osum (R)

EH (rac)

Cl

Cl

+

(R) (93%, 96% ee) Cl

A. niger EH

Figure 6.67 Convergent hydrolysis using two enzymes in sequence.

OH O

Inversion

OH (R)

(S ) Cl O (R)

Solanum tuber osum EH n tio ten e R

Cl

β-Adrenergic receptor agonists Cl (r ac) Figure 6.68 Enantioconvergent hydrolysis of m-chlorostyrene oxide using a single biocatalyst.

The enantioconvergent biohydrolysis of m-chlorostyrene oxide (Figure 6.68) in the presence of a recombinant S. tuberosum EH afforded the corresponding (R)-diol in a nearly quantitative yield [187]. The (S)-epoxide was attacked at the benzylic (more substituted) carbon whereas the (R)-epoxide was attacked at the terminal (less substituted) carbon. The racemic cis-epoxide shown in Figure 6.69 was hydrolyzed in an enantioconvergent fashion – that is, both enantiomers were converted with opposite regioselectivity (see arrows on Figure 6.69) to give the single (2S,3R)-diol, which underwent spontaneous ring closure to yield an epoxy alcohol [188]. This enzyme-triggered enantioconvergent cascade reaction was applied to the enantioselective synthesis of two natural products, (þ)-pestalotin and a Jamaican rum constituent. Cl

O

O O

O n-C 4H 9 Cl

My cobact er ium par af f inicum

HO HO HO

O

Cl

n-C 4 H9

– HCl

Jamaican rum constituent

OMe

n-C 4H 9

(+)-Pestalotin

n-C 4H 9 O

(r ac) OH

Figure 6.69 An enantioconvergent enzyme-triggered cascade reaction.

O

6.4 Hydrolysis of Epoxides

j161

O O

OH HO

Rhodococcus r uber EH

O

O

(R)

Beer aroma constituent

(r ac) Figure 6.70 Enantioconvergent hydrolysis of a trisubstituted epoxide using a single enzyme.

The enantioconvergent biohydrolysis of sterically demanding trisubstituted oxiranes has also been reported [189,190]. For instance, the enantioconvergent hydrolysis of a trisubstituted rac-epoxide (Figure 6.70) was a key step in the chemoenzymatic synthesis of a volatile constituent of the beer aroma [190]. The enantioconvergent hydrolysis of several 2,2-disubstituted epoxides was accomplished by combined bio- and chemocatalysis [191–194]. For example, the kinetic resolution of the epoxide shown in Figure 6.71 using a Nocardia sp. EH proceeded via an attack at the unsubstituted carbon atom with retention of conﬁguration at the adjacent stereogenic center [194]. In the second step, the remaining epoxide was hydrolyzed under acid-catalysis with inversion of conﬁguration. The sole ring-opening product (S)-diol was further elaborated to the natural product (R)-mevalonolactone, a biosynthetic precursor of terpenes and steroids. Figure 6.72 shows an enantioconvergent multistep process leading to an enantiopure epoxide. The racemic epoxide was resolved by A. niger EH leading to the (R)-diol and the residual (S)-epoxide with excellent optical purity [195]. The chemical O

O

H 2 SO4

OH

N ocar dia sp. EH Ph pH 7.5

O

HO OH

Ph + OH

Ph

Ph

O O (R) - Mevalonolactone

(S) (S) OH

(r ac)

Ph

Figure 6.71 Enantioconvergent hydrolysis of a 2,2-disubstituted epoxide by combined bio- and chemocatalysts. 1. NaH / THF 2. PPh 3 / CCl4 F

F O

F

A. niger EH

( r ac )

F Cl

OH

( S)

H 2O / DMSO (5%) F

+

O

Cl

Figure 6.72 An enantioconvergent process leading to an enantiopure epoxide.

(R) F

Cl OH

j 6 Transesterification and Hydrolysis of Carboxylic Acid Derivatives, Alcohols, and Epoxides

162

O

R

EH

R

OH

O

O R meso

O

O Cbz

R

N

OH ( R ,R )

O

O

Figure 6.73 EH-catalyzed desymmetrization of cyclic meso-epoxides.

O R

Halohydrin dehalogenase Nu–

OH

O R

+ (S)

R

Nu

Nu– = Cl – , CN– , NO 2– , N3 –

(R)

Figure 6.74 Ring-opening reactions catalyzed by halohydrin dehalogenase.

cyclization of the (R)-diol gave the corresponding epoxy alcohol, which was transformed into the (S)-epoxide via treatment with CCl4/PPh3. Both these reactions were shown to occur without loss of stereochemical integrity. The ﬁnal product is a key building block for the synthesis of enantiopure azole antifungal agents. meso-Epoxides are also good substrates and have the potential to yield an optically pure diol in 100% yield. The desymmetrization of a wide range of meso-epoxides was studied using high-throughput screening of environmental libraries [196]. In most cases, the (S)-stereogenic center was inverted to yield the (R,R)-diol (Figure 6.73). Moreover, the ﬁrst EHs providing access to complementary (S,S)-diols were also found. Ring opening of epoxides with nucleophiles other than water (Cl, Br, I, N3, NO2, CN) can also be catalyzed by halohydrin dehalogenase enzymes (EC 3.8.1.5, also named haloalkane dehalogenase or haloalcohol dehalogenase) (Figure 6.74) [197].

6.5 Conclusion

Organic synthesis has been revolutionized over the past two decades by the advent of enantioselective catalysis. Chemocatalysis and biocatalysis were developed in parallel and, undoubtedly, the design of many low-molecular-weight chiral catalysts was inspired by enzyme catalysis. Hydrolase-catalyzed transesteriﬁcation or hydrolysis of carboxylic acid derivatives, alcohols, and epoxides represent a versatile methodology for the preparation of optically active compounds. Hydrolases are the most frequently used biocatalysts in organic synthesis. They account for about two-thirds of all synthetic biotransformations. Several characteristics make hydrolases useful catalysts in asymmetric synthesis. They do not require sensitive cofactors and many are commercially available. They often show high stereoselectivity on various synthetic nonnatural compounds (broad substrate speciﬁcity). They are stable and active in neat organic solvents of low polarity. Unlike heavy metal catalysts, hydrolases are environmentally benign reagents (green chemistry). Recombinant protein technology

References

allows for the production of synthetically useful quantities of enzymes. Hydrolases with novel properties may be obtained by screening environmental samples. Alternatively, genetic engineering techniques such as directed evolution may lead to new biocatalysts.

Acknowledgments

The authors gratefully acknowledge the ﬁnancial support from the Natural Sciences and Engineering Research Council of Canada (NSERC).

References 1 Bornscheuer, U.T. and Kazlauskas, R.J. (2006) Hydrolases in Organic Synthesis, 2nd edn Wiley-VCH Verlag GmbH, Weinheim. 2 Buchholz, K., Kasche, V. and Bornscheuer, U.T. (2005) Biocatalysts and Enzyme Technology, Wiley-VCH Verlag GmbH, Weinheim. 3 Bommarius, A.S. and Riebel, B.R. (2004) Biocatalysis, Wiley-VCH Verlag GmbH, Weinheim. 4 Faber, K. (2004) Biotransformations in Organic Chemistry, 5th edn Springer, Berlin. 5 Drauz, K. and Waldmann, H. (eds) (2002) Enzyme Catalysis in Organic Synthesis, 2nd edn Wiley-VCH Verlag GmbH, Weinheim, Vol. 1–3. 6 Patel, R.N. (ed.) (2000) Stereoselective Biocatalysis, Marcel Dekker, New York. 7 Roberts, S.M. (ed.),(1999) Biocatalysts for Fine Chemical Synthesis, Wiley-VCH Verlag GmbH, Weinheim. 8 Secundo, F., Carrea, G., Tarabiono, C., Gatti-Lafranconi, P., Brocca, S., Lotti, M., Jaeger, K.E., Puls, M. and Eggert, T. (2006) Journal of Molecular Catalysis B: Enzymatic, 39, 166–170. 9 Hudson, E.P., Eppler, R.K. and Clark, D.S. (2005) Current Opinion in Biotechnology, 16, 637–643. 10 Secundo, F. and Carrea, G. (2003) Chemistry-A European Journal, 9, 3194–3199.

11 Reetz, M.T. (2002) Current Opinion in Chemical Biology, 6, 145–150. 12 Carrea, G. and Riva, S. (2000) Angewandte Chemie International Edition, 39, 2226–2254. 13 Klibanov, A.M. (2001) Nature, 409, 241–246. 14 Park, S. and Kazlauskas, R.J. (2003) Current Opinion in Biotechnology, 14, 432–437. 15 van Rantwijk, F., Lau, R.M. and Sheldon, R.A. (2003) Trends in Biotechnology, 21, 131–138. 16 Kragl, U., Eckstein, M. and Kaftzik, N. (2002) Current Opinion in Biotechnology, 13, 565–571. 17 Sheldon, R.A., Lau, R.M., Sorgedrager, M.J., van Rantwijk, F. and Seddon, K.R. (2002) Green Chemistry, 4, 147–151. 18 Matsuda, T., Harada, T. and Nakamura, K. (2005) Current Organic Chemistry, 9, 299–315. 19 Sakai, T., Kawabata, I., Kishimoto, T., Ema, T. and Utaka, M. (1997) Journal of Organic Chemistry, 62, 4906–4907. 20 Ema, T., Kageyama, M., Korenaga, T. and Sakai, T. (2003) Tetrahedron: Asymmetry, 14, 3943–3947. 21 Cainelli, G., Galletti, P., Giacomini, D., Gualandi, A. and Quintavalla, A. (2003) HeIvetica Chimica Acta, 86, 3548–3559. 22 Rejasse, B., Besson, T., Legoy, M.D. and Lamare, S. (2006) Organic and Biomolecular Chemistry, 4, 3703–3707. 23 Theil, F. (2000) Tetrahedron, 56, 2905–2919.

j163

j 6 Transesterification and Hydrolysis of Carboxylic Acid Derivatives, Alcohols, and Epoxides

164

24 Cao, L. (2005) Current Opinion in Chemical Biology, 9, 217–226. 25 Roy, J.J. and Abraham, T.E. (2004) Chemical Reviews, 104, 3705–3721. 26 Bornscheuer, U.T. (2003) Angewandte Chemie International Edition, 42, 3336–3337. 27 Reetz, M.T., Tielmann, P., Wiesenh€ofer, W., K€onen, W. and Zonta, A. (2003) Advanced Synthesis and Catalysis, 345, 717–728. 28 Kroutil, W., Kleewein, A. and Faber, K. (1997) Tetrahedron: Asymmetry, 8, 3251–3261. 29 For a web site, see:http://www.orgc. tugraz.at/ (by Faber, K., H€onig, H., Kleewein, A. 1994, updated 2006). 30 Sih, C.J. and Wu, S.H. (1989) Resolution of enantiomers via biocatalysis, in Topics in Stereochemistry, (eds E.L. Eliel, S.H. Wilen and N.L. Allinger), John Wiley & Sons, Ltd, New York, vol. 19, pp. 63–125. 31 Ghanem, A. and Aboul-Enein, H.Y. (2004) Tetrahedron: Asymmetry, 15, 3331–3351. 32 Robinson, D.E.J.E. and Bull, S.D. (2003) Tetrahedron: Asymmetry, 14, 1407–1446. 33 Bouzemi, N., Debbeche, H., AribiZouioueche, L. and Fiaud, J.C. (2004) Tetrahedron Letters, 45, 627–630. 34 Terao, Y., Tsuji, K., Murata, M., Achiwa, K., Nishio, T., Watanabe, N. and Seto, K. (1989) Chemical and Pharmaceutical Bulletin, 37, 1653–1655. 35 Luo, Z., Swaleh, S.M., Theil, F. and Curran, D.P. (2002) Organic Letters, 4, 2585–2587. 36 Turner, N.J. (2004) Current Opinion in Chemical Biology, 8, 114–119. 37 Pellissier, H. (2003) Tetrahedron, 59, 8291–8327. 38 Schnell, B., Faber, K. and Kroutil, W. (2003) Advanced Synthesis and Catalysis, 345, 653–666. 39 May, O., Verseck, S., Bommarius, A. and Drauz, K. (2002) Organic Process Research and Development, 6, 452–457. 40 Brown, S.A., Parker, M.C. and Turner, N.J. (2000) Tetrahedron: Asymmetry, 11, 1687–1690.

41 Um, P.J. and Drueckhammer, D.G. (1998) Journal of the American Chemical Society, 120, 5605–5610. 42 Haughton, L. and Williams, J.M.J. (2001) Synthesis, 943–946. 43 Veum, L., Kanerva, L.T., Halling, P.J., Maschmeyer, T. and Hanefeld, U. (2005) Advanced Synthesis and Catalysis, 347, 1015–1021. 44 Pesti, J.A., Yin, J., Zhang, L. and Anzalone, L. (2001) Journal of the American Chemical Society, 123, 11075–11076. 45 Boren, L., Martın-Matute, B., Xu, Y., Córdova, A. and B€ackvall, J.E. (2006) Chemistry-A European Journal, 12, 225–232. 46 Choi, J.H., Choi, Y.K., Kim, Y.H., Park, E.S., Kim, E.J., Kim, M.J. and Park, J. (2004) Journal of Organic Chemistry, 69, 1972–1977. 47 Pamies, O. and B€ackvall, J.E. (2003) Chemical Reviews, 103, 3247–3261. 48 Kim, M.J., Ahn, Y. and Park, J. (2002) Current Opinion in Biotechnology, 13, 578–587. 49 Garcia-Urdiales, E., Alfonso, I. and Gotor, V. (2005) Chemical Reviews, 105, 313–354. 50 Schoffers, E., Golebiowski, A. and Johnson, C.R. (1996) Tetrahedron, 52, 3769–3826. 51 Faber, K. (2001) Chemistry-A European Journal, 7, 5004–5010. 52 Leue, S., Miao, W., Kanazawa, A., Genisson, Y., Garçon, S. and Greene, A.E. (2001) Journal of the Chemical Society, Perkin Transactions 1, 2903–2905. 53 Truppo, M.D., Journet, M., Shaﬁee, A. and Moore, J.C. (2006) Organic Process Research and Development, 10, 592–598. 54 Chikusa, Y., Hirayama, Y., Ikunaka, M., Inoue, T., Kamiyama, S., Moriwaki, M., Nishimoto, Y., Nomoto, F., Ogawa, K., Ohno, T., Otsuka, K., Sakota, A.K., Shirasaka, N., Uzura, A. and Uzura, K. (2003) Organic Process Research and Development, 7, 289–296. 55 Salgado, A., Huybrechts, T., Eeckhaut, A., Van der Eycken, J., Szakonyi, Z., F€ ul€op, F.,

References

56 57

58

59

60

61

62

63

64

65

66

67

68

69

70 71

Tkachev, A. and De Kimpe, N. (2001) Tetrahedron, 57, 2781–2786. Kedrowski, B.L. (2003) Journal of Organic Chemistry, 68, 5403–5406. Back, T.G. and Wulff, J.E. (2004) Angewandte Chemie International Edition, 43, 6493–6496. White, J.D., Kim, J. and Drapela, N.E. (2000) Journal of the American Chemical Society, 122, 8665–8671. Katoh, T., Kakiya, K., Nakai, T., Nakamura, S., Nishide, K. and Node, M. (2002) Tetrahedron: Asymmetry, 13, 2351–2358. Liang, X., Lohse, A. and Bols, M. (2000) Journal of Organic Chemistry, 65, 7432–7437. Henke, E., Schuster, S., Yang, H. and Bornscheuer, U.T. (2000) Monatshefte fur Chemie, 131, 633–638. Liljeblad, A., Lindborg, J. and Kanerva, L.T. (2000) Tetrahedron: Asymmetry, 11, 3957–3966. Miyazawa, T., Onishi, K., Murashima, T., Yamada, T. and Tsai, S.W. (2005) Tetrahedron: Asymmetry, 16, 2569–2573. Kawasaki, M., Kakuda, H., Goto, M., Kawabata, S. and Kometani, T. (2003) Tetrahedron: Asymmetry, 14, 1529–1534. Morrone, R., Piattelli, M. and Nicolosi, G. (2001) European Journal of Organic Chemistry, 1441–1443. Hedenstr€ om, E., Nguyen, B.V. and Silks, L.A. (2002) Tetrahedron: Asymmetry, 13, 835–844. Enzelberger, M.M., Bornscheuer, U.T., Gatﬁeld, I. and Schmid, R.D. (1997) Journal of Biotechnology, 56, 129–133. Sakai, N., Ageishi, S., Isobe, H., Hayashi, Y. and Yamamoto, Y. (2000) Journal of the Chemical Society, Perkin Transactions, 1, 71–77. Klomp, D., Peters, J.A. and Hanefeld, U. (2005) Tetrahedron: Asymmetry, 16, 3892–3896. Mori, K. (1995) Synlett, 1097–1109. Missio, L.J. and Comasseto, J.V. (2000) Tetrahedron: Asymmetry, 11, 4609–4615.

72 Jeon, N.Y., Ko, S.J., Won, K., Kang, H.Y., Kim, B.T., Lee, Y.S. and Lee, H. (2006) Tetrahedron Letters, 47, 6517–6520. 73 Kumar, A., Gross, R.A. and Jendrossek, D. (2000) Journal of Organic Chemistry, 65, 7800–7806. 74 Hwang, B.Y., Lee, H.B., Kim, Y.G. and Kim, B.G. (2006) Journal of Molecular Catalysis B: Enzymatic, 41, 125–129. 75 Ozegowski, R., Kunath, A. and Schick, H. (1995) Tetrahedron: Asymmetry, 6, 1191–1194. 76 Fryszkowska, A., Komar, M., Koszelewski, D. and Ostaszewski, R. (2006) Tetrahedron: Asymmetry, 17, 961–966. 77 Fryszkowska, A., Frelek, J. and Ostaszewski, R. (2005) Tetrahedron, 61, 6064–6072. 78 Wang, M.X. (2005) Topics in Catalysis, 35, 117–130. 79 Martinkova, L. and Kren, V. (2002) Biocatalysis and Biotransformation, 20, 73–93. 80 Banerjee, A., Sharma, R. and Banerjee, U.C. (2002) Applied Microbiology and Biotechnology, 60, 33–44. 81 Kobayashi, M. and Shimizu, S. (2000) Current Opinion in Chemical Biology, 4, 95–102. 82 Nagasawa, T. and Yamada, H. (1995) Pure and Applied Chemistry, 67, 1241–1256. 83 Wang, M.X. and Zhao, S.M. (2002) Tetrahedron Letters, 43, 6617–6620. 84 Wang, M.X., Lin, S.J., Liu, C.S., Zheng, Q.Y. and Li, J.S. (2003) Journal of Organic Chemistry, 68, 4570–4573. 85 Wang, M.X. and Feng, G.Q. (2003) Journal of Organic Chemistry, 68, 621–624. 86 Mateo, C., Chmura, A., Rustler, S., van Rantwijk, F., Stolz, A. and Sheldon, R.A. (2006) Tetrahedron: Asymmetry, 17, 320–323. 87 DeSantis, G., Zhu, Z., Greenberg, W.A., Wong, K., Chaplin, J., Hanson, S.R., Farwell, B., Nicholson, L.W., Rand, C.L., Weiner, D.P., Robertson, D.E. and Burk, M.J. (2002) Journal of the American Chemical Society, 124, 9024–9025.

j165

j 6 Transesterification and Hydrolysis of Carboxylic Acid Derivatives, Alcohols, and Epoxides

166

88 Gr€oger, H. (2001) Advanced Synthesis and Catalysis, 343, 547–558. 89 Yamamoto, K., Oishi, K., Fujimatsu, I. and Komatsu, K.I. (1991) Applied and Environment Microbiology, 57, 3028–3032. 90 Wang, M.X., Lin, S.J., Liu, J. and Zheng, Q.Y. (2004) Advanced Synthesis and Catalysis, 346, 439–445. 91 Wegman, M.A., Heinemann, U., van Rantwijk, F., Stolz, A. and Sheldon, R.A. (2001) Journal of Molecular Catalysis B: Enzymatic, 11, 249–253. 92 DeSantis, G., Wong, K., Farwell, B., Chatman, K., Zhu, Z., Tomlinson, G., Huang, H., Tan, X., Bibbs, L., Chen, P., Kretz, K. and Burk, M.J. (2003) Journal of the American Chemical Society, 125, 11476–11477. 93 Kakeya, H., Sakai, N., Sano, A., Yokoyama, M., Sugai, T. and Ohta, H. (1991) Chemistry Letters, 20, 1823–1824. 94 Bergeron, S., Chaplin, D.A., Edwards, J.H., Ellis, B.S.W., Hill, C.L., Holt-Tifﬁn, K., Knight, J.R., Mahoney, T., Osborne, A.P. and Ruecroft, G. (2006) Organic Process Research and Development, 10, 661–665. 95 M€ uller, M. (2005) Angewandte Chemie International Edition, 44, 362–365. 96 Wu, Z.L. and Li, Z.Y. (2003) Chemical Communications, 386–387. 97 Liljeblad, A. and Kanerva, L.T. (2006) Tetrahedron, 62, 5831–5854. 98 Shaw, N.M., Robins, K.T. and Kiener, A. (2003) Advanced Synthesis and Catalysis, 345, 425–435. 99 Peterson, M. and Sauter, M. (1999) Chimia, 53, 608–612. 100 Braiuca, P., Boscarol, L., Ebert, C., Linda, P. and Gardossi, L. (2006) Advanced Synthesis and Catalysis, 348, 773–780. 101 Sonke, T., Kaptein, B., Wagner, A.F.V., Quaedﬂieg, P.J.L.M., Schultz, S., Ernste, S., Schepers, A., Mommers, J.H.M. and Broxterman, Q.B. (2004) Journal of Molecular Catalysis B: Enzymatic, 29, 265–277. 102 Sonke, T., Ernste, S., Tandler, R.T., Kaptein, B., Peeters, W.P.H., van Assema,

103

104 105

106

107

108 109

110 111

112

113 114

115

116

117

F.B.J., Wubbolts, M.G. and Schoemaker, H.E. (2005) Applied and Environment Microbiology, 71, 7961–7973. Asano, Y. and Yamaguchi, S. (2005) Journal of the American Chemical Society, 127, 7696–7697. Shaw, N.M. and Naughton, A.B. (2004) Tetrahedron, 60, 747–752. Shaw, N.M., Naughton, A., Robins, K., Tinschert, A., Schmid, E., Hischier, M.L., Venetz, V., Werlen, J., Zimmermann, T., Brieden, W., de Riedmatten, P., Roduit, J.P., Zimmermann, B. and Neum€ uller, R. (2002) Organic Process Research and Development, 6, 497–504. Lloyd, R.C., Lloyd, M.C., Smith, M.E.B., Holt, K.E., Swift, J.P., Keene, P.A., Taylor, S.J.C. and McCague, R. (2004) Tetrahedron, 60, 717–728. Taylor, S.J.C., Brown, R.C., Keene, P.A. and Taylor, I.N. (1999) Bioorganic and Medicinal Chemistry, 7, 2163–2168. Forró, E. and F€ ul€op, F. (2006) Chemistry-A European Journal, 12, 2587–2592. Forró E., Paal, T., Tasnadi, G. and F€ ul€op, F. (2006) Advanced Synthesis and Catalysis, 348, 917–923. Forró, E. and F€ ul€op, F. (2003) Organic Letters, 5, 1209–1212. Park, S., Forró E., Grewal, H., F€ ul€op, F. and Kazlauskas, R.J. (2003) Advanced Synthesis and Catalysis, 345, 986–995. Adam, W., Groer, P., Humpf, H.U. and Saha-M€oller, C.R. (2000) Journal of Organic Chemistry, 65, 4912–4922. Nam, S.H., Park, H.S. and Kim, H.S. (2005) Chemical Record, 5, 298–307. Burton, S.G. and Dovington, R.A. (2004) Tetrahedron: Asymmetry, 15, 2737–2741. Altenbuchner, J., Siemann-Herzberg, M. and Syldatk, C. (2001) Current Opinion in Biotechnology, 12, 559–563. Ch^enevert, R., Pelchat, N. and Jacques, F. (2006) Current Organic Chemistry, 10, 1067–1094. Santaniello, E., Casati, S. and Ciuffreda, P. (2006) Current Organic Chemistry, 10, 1095–1123.

References 118 Gotor-Fernandez, V., Busto, E. and Gotor, V. (2006) Advanced Synthesis and Catalysis, 348, 797–812. 119 Lutz, S. (2004) Tetrahedron: Asymmetry, 15, 2743–2748. 120 Kirk, O. and Christensen, M.W. (2002) Organic Process Research and Development, 6, 446–451. 121 Dominguez de Maria, P., SanchezMontero, J.M., Sinisterra, J.V. and Alcantara, A.R. (2006) Biotechnology Advances, 24, 180–196. 122 Dominguez de Maria, P., Alcantara, A.R., Carballeira, J.D., de la Casa, R.M., Garcia-Burgos, C.A., Hernaiz, M.J., Sanchez-Montero, J.M. and Sinisterra, J.V. (2006) Current Organic Chemistry, 10, 1053–1066. 123 Hanefeld, U. (2003) Organic and Biomolecular Chemistry, 1, 2405–2415. 124 Otera, J. (2003) Esteriﬁcation Methods, Reactions and Applications, Wiley-VCH Verlag GmbH, Weinheim. 125 Kita, Y. and Akai, S. (2004) The Chemical Record, 4, 363–372. 126 Izquierdo, I., Plaza, M.T., Rodriguez, M. and Tamayo, J. (1999) Tetrahedron: Asymmetry, 10, 449–455. 127 Ch^enevert, R., Goupil, D., Rose, Y.S. and Bedard, E. (1998) Tetrahedron: Asymmetry, 9, 4285–4288. 128 Kadota, K., ElAzab, A.S., Taniguchi, T. and Ogasawara, K. (2000) Synthesis, 10, 1372–1374. 129 Lee, T. and Kim, S. (2003) Tetrahedron: Asymmetry, 14, 1951–1954. 130 Costa, C.E., Cososki, G.C., Barchesi, H.B., Zanotto, S.P., Nascimento, M.G. and Comasseto, J.V. (2004) Tetrahedron: Asymmetry, 15, 3945–3954. 131 Dos Santos, A.A., Da Costa, C.E., Princival, J.L. and Comasseto, J.V. (2006) Tetrahedron: Asymmetry, 17, 2252–2259. 132 Wallner, S.R., Nestl, B.M. and Faber, K. (2005) Organic and Biomolecular Chemistry, 3, 2652–2656. 133 Gotor-Fernandez, V., Brieva, R. and Gotor, V. (2006) Journal of Molecular Catalysis B: Enzymatic, 40, 111–120.

134 Breuer, M., Ditrich, K., Habicher, T., Hauer, B., Keßeler, M., St€ urmer, R. and Zelinski, T. (2004) Angewandte Chemie International Edition, 43, 788–824. 135 Garcia-Junceda, E., Garcia-Garcia, J.F., Bastida, A. and Fernandez-Mayoralas, A. (2004) Bioorganic and Medicinal Chemistry, 12, 1817–1834. 136 Straathof, A.J.J., Panke, S. and Schmid, A. (2002) Current Opinion in Biotechnology, 13, 548–556. 137 Zaks, A. and Dodds, D.R. (1997) Drug Discovery Today, 2, 513–531. 138 Ikunaka, M. (2004) Catalysis Today, 96, 93–102. 139 Nakagawa, H., Sugahara, T. and Ogasawara, K. (2000) Organic Letters, 2, 3181–3183. 140 Morita, Y. Tokuyama, H. and Fukuyama, T. (2005) Organic Letters, 7, 4337–4340. 141 Lambusta, D., Nicolosi, G., Patti, A. and Sanﬁlippo, C. (2003) Journal of Molecular Catalysis B: Enzymatic, 22, 271–277. 142 Alibes, R., Bayón, P., de March, P., Figueredo, M., Font, J. and Marjanet, G. (2006) Organic Letters, 8, 1617–1620. 143 Akai, S. and Tanimoto, K. and Kita, Y. (2004) Angewandte Chemie International Edition, 43, 1407–1410. 144 Akai, S., Naka, T., Omura, S., Tanimoto, K., Imanishi, M., Takebe, Y., Matsugi, M. and Kita, Y. (2002) Chemistry-A European Journal, 8, 4255–4264. 145 Akai, S., Tanimoto, K., Kanao, Y., Omura, S. and Kita, Y. (2005) Chemical Communications, 2369–2371. 146 Fujii, M., Fukumura, M., Hori, Y., Hirai, Y., Akita, H., Nakamura, K., Toriizuka, K. and Ida, Y. (2006) Tetrahedron: Asymmetry, 17, 2292–2298. 147 Gruttadauria, M., Lo Meo, P. and Noto, R. (2004) Tetrahedron Letters, 45, 83–85. 148 Sundby, E., Perk, L., Anthonsen, T., Aasen, A.J. and Hansen, T.V. (2004) Tetrahedron, 60, 521–524. 149 Sano, S., Miwa, T., Hayashi, K., Nozaki, K., Ozaki, Y. and Nagao, Y. (2001) Tetrahedron Letters, 42, 4029–4031.

j167

j 6 Transesterification and Hydrolysis of Carboxylic Acid Derivatives, Alcohols, and Epoxides

168

150 Anstiss, M., Holland, J.M., Nelson, A. and Titchmarsh, J.R. (2003) Synlett, 1213–1220. 151 B€ ohm, C., Austin, W.F. and Trauner, D. (2003) Tetrahedron: Asymmetry, 14, 71–74. 152 Harada, T. and Oku, A. (1994) Synlett, 95–104. 153 Ch^enevert, R., Courchesne, G. and Caron, D. (2003) Tetrahedron: Asymmetry, 14, 2567–2571. 154 Ch^enevert, R. and Rose, Y.S. (2000) Journal of Organic Chemistry, 65, 1707–1709. 155 Fellows, I.M., Kaelin, D.E. and Martin, S.F. (2000) Journal of the American Chemical Society, 122, 10781–10787. 156 Ch^enevert, R., Courchesne, G. and Pelchat, N. (2006) Bioorganic and Medicinal Chemistry, 14, 5389–5396. 157 Berkowitz, D.B., Choi, S. and Maeng, J.H. (2000) Journal of Organic Chemistry, 65, 847–860. 158 Mehta, G. and Islam, K. (2004) Organic Letters, 6, 807–810. 159 Konno, H. and Ogasawara, K. (1999) Synthesis, 1135–1140. 160 Donohoe, T.J., Johnson, P.D., Pye, R.J. and Keenan, M. (2005) Organic Letters, 7, 1275–1277. 161 Tanaka, H., Kozuki, Y. and Ogasawara, K. (2002) Tetrahedron Letters, 43, 4175–4178. 162 Mehta, G. and Islam, K. (2004) Tetrahedron Letters, 45, 7683–7687. 163 Yudin A.K. (ed.) (2006) Aziridines and Epoxides in Organic Synthesis, Wiley-VCH Verlag GmbH, Weinheim. 164 Xia, Q.H., Ge, H.Q., Ye, C.P., Liu, Z.M. and Su, K.X. (2005) Chemical Reviews, 105, 1603–1662. 165 Bernhardt, R. (2006) Journal of Biotechnology, 124, 128–145. 166 Meunier, B. de Visser, S.P. and Shaik, S. (2004) Chemical Reviews, 104, 3947–3980. 167 Dembitsky, V.M. (2003) Tetrahedron, 59, 4701–4720. 168 Jacobsen, E.N. (2000) Accounts of Chemical Research, 33, 421–431. 169 Smit, M.S. and Labuschagne, M. (2006) Current Organic Chemistry, 10, 1145–1161.

170 van Loo, B., Kingma, J., Arand, M., Wubbolts, M.G. and Janssen, D.B. (2006) Applied and Environment Microbiology, 72, 2905–2917. 171 Smit, M.S. (2004) Trends in Biotechnology, 22, 123–129. 172 de Vries, E.J. and Janssen, D.B. (2003) Current Opinion in Biotechnology, 14, 414–420. 173 Archelas, A. and Furstoss, R. (2001) Current Opinion in Chemical Biology, 5, 112–119. 174 Steinreiber, A. and Faber, K. (2001) Current Opinion in Biotechnology, 12, 552–558. 175 Kim, H.S., Lee, O.K., Lee, S.J., Hwang, S., Kim, S.J., Yang, S.H., Park, S. and Lee, E.Y. (2006) Journal of Molecular Catalysis B: Enzymatic, 41, 130–135. 176 Xu, W., Xu, J.H., Pan, J., Gu, Q. and Wu, X.Y. (2006) Organic Letters, 8, 1737–1740. 177 Belder, D., Ludwig, M., Wang, L.W. and Reetz, M.T. (2006) Angewandte Chemie International Edition, 45, 2463–2466. 178 Reetz, M.T., Wang, L.W. and Bocola, M. (2006) Angewandte Chemie International Edition, 45, 1236–1241. 179 Cedrone, F., Bhatnagar, T. and Baratti, J.C. (2005) Biotechnology Letters, 27, 1921–1927. 180 Karboune, S., Archelas, A. and Baratti, J. (2006) Enzyme and Microbial Technology, 39, 318–324. 181 Hopmann, K.H. and Himo, F. (2006) Chemistry-A European Journal, 12, 6898–6909. 182 Hopmann, K.H., Hallberg, B.M. and Himo, F. (2005) Journal of the American Chemical Society, 127, 14339–14347. 183 Genzel, Y., Archelas, A., Broxterman, Q.B., Schulze, B. and Furstoss, R. (2001) Journal of Organic Chemistry, 66, 538–543. 184 Cleij, M., Archelas, A. and Furstoss, R. (1999) Journal of Organic Chemistry, 64, 5029–5035. 185 Pedragosa-Moreau, S., Archelas, A. and Furstoss, R. (1993) Journal of Organic Chemistry, 58, 5533–5536. 186 Manaj, K.M., Archelas, A., Baratti, J. and Furstoss, R. (2001) Tetrahedron, 57, 695–701.

Further Reading 187 Monterde, M.I., Lombard, M., Archelas, A., Cronin, A., Arand, M. and Furstoss, R. (2004) Tetrahedron: Asymmetry, 15, 2801–2805. 188 Mayer, S.F., Steinreiber, A., Goriup, M., Saf, R. and Faber, K. (2002) Tetrahedron: Asymmetry, 13, 523–528. 189 Steinreiber, A., Mayer, S.F., Saf, R. and Faber, K. (2001) Tetrahedron: Asymmetry, 12, 1519–1528. 190 Steinreiber, A., Mayer, S.F. and Faber, K. (2001) Synthesis, 13, 2035–2039. 191 Simeo, Y. and Faber, K. (2006) Tetrahedron: Asymmetry, 17, 402–409. 192 Steinreiber, A., Hellstr€om, M., Mayer, S.F., Orru, R.V.A. and Faber, K. (2001) Synlett, 111–113. 193 Orru, R.V.A., Mayer, S.F., Kroutil, W. and Faber, K. (1998) Tetrahedron, 54, 859–874. 194 Orru, R.V.A., Osprian, I., Kroutil, W. and Faber, K. (1998) Synthesis, 1259–1263. 195 Monfort, N., Archelas, A. and Furstoss, R. (2004) Tetrahedron, 60, 601–605.

196 Zhao, L., Han, B., Huang, Z., Miller, M., Huang, H., Malashock, D.S., Zhu, Z., Milan, A., Robertson, D.E., Weiner, D.P. and Burk, M.J. (2004) Journal of the American Chemical Society, 126, 11156–11157. 197 Elenkov, M.M., Tang, L., Hauer, B. and Janssen, D.B. (2006) Organic Letters, 8, 4229–4257.

Further Reading Cardillo, G., Gentilucci, L., Tolomelli, A. and Tomasini, C. (1998) Journal of Organic Chemistry, 63, 2351–2353. Cardillo, G. and Tolomelli, A. and Tomasini, C. (1996) Journal of Organic Chemistry, 61, 8651–8654. Fadnavis, N.W., Radhika, K.R. and Devi, A.V. (2006) Tetrahedron: Asymmetry, 17, 240–244. Topgi, R.S., Ng, J.S., Landis, B., Wang, P. and Behling, J.R. (1999) Bioorganic and Medicinal Chemistry, 7, 2221–2229.

j169

j171

7 Aminolysis and Ammonolysis of Carboxylic Acid Derivatives Vicente Gotor-Fernandez and Vicente Gotor

7.1 Introduction

The use of enzymes for organic synthesis has become an interesting area for organic and bioorganic chemists, since many enzymes have demonstrated their activity with nonnatural substrates in organic media, allowing the development of many different synthetic transformations. Among all of them, hydrolases are the most used enzymes owing to their high stability and activity with a broad spectrum of substrates. Besides, many of them are commercially available, do not need cofactor, and work under mild reaction conditions in contrast with drastic reaction conditions required for chemical processes. From the hydrolases, lipases (EC 3.1.1.3) are the most popular biocatalysts and nowadays the ones more used in asymmetric synthesis. Applications of lipases include kinetic resolution (KR) of racemic alcohols, carboxylic acids, esters, or amines [1], as well as the desymmetrization of prochiral compounds [2]. Moreover, lipases are ideal biocatalysts for ammonolysis and aminolysis of esters because they present very low amidase activity. Recently, it has been reported that amidase activity of different benzylamides depends on the acyl residue and also of the substituent groups in the aromatic ring [3]. Over the past years, interest in the preparation of chiral amines and amides by enzymatic ammonolysis or aminolysis reactions [4] has greatly increased for academic and industrial sectors. The role that the enzymatic acylation of amines or ammonia plays for the preparation of some pharmaceuticals is noteworthy [5]. Although the aminolysis of esters to amides is a useful synthetic operation, usually it presents some disadvantages in terms of drastic reaction conditions, long reaction times or strong alkali metal as catalyst, which are usually not compatible with other functional groups in the molecule [6]. For this reason, enzymatic aminolysis of carboxylic acid derivatives offers a clean and ecological way for the preparation of different kind of amines and amides in a regio-, chemo-, and enantioselective manner. In the last decade, biocatalysis in nonaqueous media, using hydrolases, has been widely used for organic chemists. The possibilities that these biocatalysts offer for the preparation of different types of organic compounds, depending upon the nucleophile Asymmetric Organic Synthesis with Enzymes. Edited by Vicente Gotor, Ignacio Alfonso, and Eduardo Garcıa-Urdiales Copyright Ó 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 978-3-527-31825-4

j 7 Aminolysis and Ammonolysis of Carboxylic Acid Derivatives

172

O 1

R

OR

O 1

O 2

R OR Transesterification

O H2O2

R OOH Perhydrolysis

R

R OH Hydrolysis

H2O

NHR

Aminolysis O

NH3

Enz

Acyl–enzyme complex O

1

1

2

R

R NH2

2

R OH

O 1

1

2

1

R NH2 Ammonolysis

2

R NHNH2

O

1

2 R NHNHR Hydrazynolysis

Scheme 7.1 Biotransformations of carboxylic acid derivatives using lipases.

used in the process, is shown in Scheme 7.1. The use of Candida antarctica lipase B (CALB) in enzymatic ammonolysis and aminolysis reactions [7] has shown a great utility for the synthesis of a great variety of chiral nitrogenated organic compounds. In this chapter, the desire of the authors is to show a range of examples that cover the different possibilities that hydrolytic enzymes, specially lipases, can offer in the preparation of nitrogenated organic compounds through enzymatic ammonolysis and aminolysis of carboxylic acid derivatives, and thus to present an easily accessible bibliographic comprehension of this subject. Although, transesteriﬁcation processes have received major attention, in the last few years, the enzymatic acylation of amines for synthetic purposes is being employed as a conventional tool for the synthesis of chiral amines and amides; however, the preparation of other nitrogenated compounds, such as hydrazides, by enzymatic hydrazinolysis reaction, is a less practical process and it has been much less investigated [8].

7.2 Mechanism of Enzymatic Ammonolysis and Aminolysis Reactions

The mechanism for the lipase-catalyzed reaction of an acid derivative with a nucleophile (alcohol, amine, or thiol) is known as a serine hydrolase mechanism (Scheme 7.2). The active site of the enzyme is constituted by a catalytic triad (serine, aspartic, and histidine residues). The serine residue accepts the acyl group of the ester, leading to an acyl–enzyme activated intermediate. This acyl–enzyme intermediate reacts with the nucleophile, an amine or ammonia in this case, to yield the ﬁnal amide product and leading to the free biocatalyst, which can enter again into the catalytic cycle. A histidine residue, activated by an aspartate side chain, is responsible for the proton transference necessary for the catalysis. Another important factor is that the oxyanion hole, formed by different residues, is able to stabilize the negatively charged oxygen present in both the transition state and the tetrahedral intermediate.

Scheme 7.2 Mechanism of serine hydrolases.

7.2 Mechanism of Enzymatic Ammonolysis and Aminolysis Reactions

j173

j 7 Aminolysis and Ammonolysis of Carboxylic Acid Derivatives

174

The complex structure of the enzyme can show a very large substrate–enzyme interaction speciﬁcity, which can be traduced to a high degree of chemo-, regio-, or stereoselectivity. For this reason, nowadays, the versatility of biotransformations for synthetic proposals is an excellent tool for organic chemists [9].

7.3 Aminolysis and Ammonolysis of Carboxylic Acids

Although, the enzymatic reaction of esters with amines or ammonia have been well documented, the corresponding aminolysis with carboxylic acids are rarer, because of the tendency of the reactants to form unreactive salts. For this reason some different strategies have been used to avoid this problem. Normally, this reaction has been used for the preparation of amides of industrial interest, for instance, one of the most important amides used in the polymer industry like oleamide has been produced by enzymatic amidation of oleic acid with ammonia and CALB in different organic solvents [10]. To carry out the enzymatic amidation of carboxylic acids, normally two strategies are considered: the use of ionic liquids or the removal of water from the reaction media at high temperature or reduced pressure. For instance, one of the ﬁrst examples of the use of ionic liquids in biocatalysis has been the preparation of octanamide from octanoic acid as starting material and ammonia in the presence of CALB (Scheme 7.3) [11]. Irimescu and Kato have recently described an interesting example of enzymatic KR in ionic liquids instead of organic solvents (Scheme 7.4) [12]. The resolution with CALB is based on the fact that the reaction equilibrium was shifted toward the amide synthesis by the removal of water under reduced pressure. Nonsolvent systems have been also employed in this enantioselective amidation processes, reacting racemic amines with aliphatic acids. The best reaction conditions for the conversion of acids to amides was observed using CALB at 90 C under vacuum. Meanwhile, no O OH

NH3 CALB [C4mim][BF4] 40 ºC, 4 days

O NH2 Quantitative conversion

Scheme 7.3 Example of an ammonolysis process using an ionic liquid-like solvent.

Me Ph

NH2 or

+

Me Ph

Me

Me

NH2

COOH

+

Vacuum CALB Ionic liquid

Ph

NH2

Ph

O

N H

or Me

Me Ph

NH2

Scheme 7.4 Enzymatic acylation of amines with carboxylic acids.

+ Ph

H N O

7.3 Aminolysis and Ammonolysis of Carboxylic Acids O

Ph

O NH2

+ HO R

Ph

j175

NHX 1) Thermolisin KPi buffer pH 7.5 2) TFA: H2O (95:5)

O HO O

N H

NHX R

X= Fmoc, Cbz R= H, CH2CH(CH3)2, CH2Ph, CH2CH2CONH2, + – CH2OH, CH2(imidazole)H , CH2COO , CH2CH2CH2CH3

Scheme 7.5 Enzymatic synthesis of dipeptides on solid phase.

amidation occurred in the absence of enzyme or in the presence of organic solvents, and this methodology has been employed to catalyze the enantioselective amidation of aliphatic acids by racemic 2-ethylhexyl amine [13]. Although, there are not many examples using lipases with solid phase substrates, some processes have been described. The direct enzymatic synthesis of peptides from aminoacids is an interesting alternative to chemical synthesis. The preparation of some dipeptidesonsolidsupporthasresultedinaveryusefulprocess;inthiscasethermolysin, a protease, is more efﬁcient than lipases (Scheme 7.5) [14]. The amine is supported on PEGA [poly-(ethylene glycol)-acrylamide] and owing to the positive charge of the resin, which avoids the ionization of the amine, it is possible to achieve good yields. Alkanolamides from fatty acids are environmentally benign surfactants useful in a wide range of applications. It was found that most lipases catalyze both amidation and the esteriﬁcation of alkanolamides; however, normally the predominant ﬁnal products are the corresponding amides, via amidation, and also by esteriﬁcation and subsequent migration [15]. Recently, an interesting example for the production of novel hydroxylated fatty amides in organic solvents has been carried out by Kuo et al. [16]. Also the impact of various reaction parameters on enzymatic synthesis of amide surfactants from ethanolamine and diethanolamine has been studied, although the possibilities of acyl migration are not investigated. However, it was found that the selectivity of the reaction depended on the solubility of the product in the solvent used, and that the choice of solvent was critical to obtain an efﬁcient process [17]. Recently, an environmentally benign and volume efﬁcient process for enzymatic production of alkanolamides has been described where CALB catalyzes the amidation of lauric acid and ethanolamine in the absence of solvent, at 90 C, to keep the reactants in a liquid state and to remove the water [18]. The enzyme was both very active and stable under the reaction conditions, with about half of the activity remaining after two weeks, obtaining the ﬁnal amide with a 95% yield (Scheme 7.6). O

O OH

Lipase O

NH2

+ H2N

OH

Lipase O N H

Scheme 7.6 Solvent-free enzymatic synthesis of fatty alkanolamines.

OH

j 7 Aminolysis and Ammonolysis of Carboxylic Acid Derivatives

176

7.4 Aminolysis and Ammonolysis of Esters

The aminolysis of esters has been much more investigated than the corresponding aminolysis of carboxylic acids. This process is of great utility for the synthesis of nitrogenated organic compounds, especially for the resolution of amines and diamines; moreover, in some cases this process is useful for resolution of esters and for the preparation of amides of industrial interest in benign conditions. Recently, the possibilities of this reaction for the preparation of highly enantiopure b-aminoacids have also been reviewed [19]. In this section, we brieﬂy comment on some representative examples of the enzymatic aminolysis and ammonolysis of esters for the preparation of nitrogenated compounds that are difﬁcult to obtain by conventional chemical methods.

7.4.1 Preparation of Nonchiral Amides

The lipase-catalyzed aminolysis of nonracemic esters has been an excellent methodology for the preparation of amides, which are difﬁcult to obtain by chemical procedures. In this section, the preparation of fatty amides and compounds used in a wide variety of industrial applications is particularly relevant. One of the ﬁrst processes to obtain oleamide has been carried out by enzymatic ammonolysis reaction of triolein and ammonia. Again, CALB is the most efﬁcient catalyst; this procedure has the advantage that puriﬁcation steps are not necessary [20]. A practical and chemoselective example of enzymatic aminolysis reaction is the preparation of N-(4-methyl-4-azapentyl) acrylamide (Scheme 7.7), an important building block in the preparation of polymers. A comprehensive study of all possible parameters affecting the enzymatic aminolysis of methyl acrylate and N,N-dimethyl-1, 3-propanediamine and CALB showed that a good selectivity was achieved when the addition of the amine was done by continuous ﬂow addition during long reaction times. Under these conditions, the formation of the Michael adduct is minimized [21]. An efﬁcient biocatalytic method for the production of amides in multigram scale has been developed for the synthesis of a pyrrole-amide, which is an intermediate for the synthesis of the dipeptidyl peptidase IV that regulates plasma levels of the insulinotropic proglucagon. CALB catalyzes the ammonolysis of the ester with ammonium carbamate as source of ammonia (Scheme 7.8) [22]. The use of ascarite and calcium chloride as adsorbents for carbon dioxide and ethanol by-products,

OMe O

+ H2N

NMe2

CAL-B 1,4-Dioxane

H N O

Scheme 7.7 Chemoselective synthesis of an acrylamide.

NMe2

+

H N

MeO O

NMe2

7.4 Aminolysis and Ammonolysis of Esters O N OEt O

O

CALB NH2CO2NH4 CaCl2, Ascarite Toluene 50 ºC, 3 days

O

j177

N NH2 O

O

81% isolated yield

Scheme 7.8 Enzymatic ammonolysis using ammonium carbamate as source of ammonia.

respectively, increases the yield to 98% overcoming traditional chemical routes, which yield just 64% of the desired product. Surprisingly, the p-system geometry in a substrate has a notable inﬂuence in the enzymatic aminolysis of esters. The reaction of diethyl fumarate with different amines or ammonia in the presence of CALB led to the corresponding transamidoesters with good isolated yields, but in the absence of enzyme, a high percentage of the corresponding Michael adduct is obtained (Scheme 7.9). Enzymatic aminolysis of diethyl maleate led to the recovery of the same a,b-unsaturated amidoester, diethyl fumarate, and diethyl maleate. The explanation of these results can be rationalized via a previous Michael/retro-Michael type isomerization of diethyl maleate to fumarate, before the enzymatic reaction takes place. In conclusion, diethylmaleate is not an adequate substrate for this enzymatic aminolysis reaction [23]. Recently Lin and coworkers have developed a selective synthesis of N-acyl and O-acyl propanolol vinyl derivatives by enzyme-catalyzed acylation of propanolol using divinyl dicarboxylates with different carbon chain lengths (Scheme 7.10) [24]. Lipase AY30 in diisopropyl ether demonstrated high chemoselectivity toward the amino

O O OEt EtO

O

NHR OEt

+ EtO O

OEt EtO

O

+ RNH2

O

O

CALB CALB

O

O NHR

EtO

+

N H

EtO

O OEt

O ( )n

N H

O

OEt O

+ RNH2

OEt

O

O

O

O

OEt EtO

OEt

+

OEt O

Scheme 7.9 Enzymatic aminolysis of a,b-unsaturated diesters using CALB.

O

NHR OEt

+ EtO O

j 7 Aminolysis and Ammonolysis of Carboxylic Acid Derivatives

178

O Lipase AY30 i Pr2O

N OH O

( )n O O

O OH

N H

+

COOCHCH2 ( )n COOCHCH2

MJL 1,4-Dioxane

n = 2, 4, 8 O O O

N H ( )n O O

Scheme 7.10 Controllable selective synthesis of N-acyl or O-accylpropanolol vinyl esters catalyzed by lipases.

group, whereas lipase M from Mucor javanicus in 1,4-dioxane selectively acylated the hydroxyl function. 7.4.2 Resolution of Esters

The enzymatic resolution of esters via aminolysis or ammonolysis processes represents an efﬁcient alternative to the resolution of substrates by transesteriﬁcation and hydrolysis processes. Of special relevance is the resolution of bifunctional compounds – for instance, an interesting process is the enantiopure preparation of b-hydroxyesters and b-hydroxyamides by enzymatic resolution of 3-hydroxyesters [25]. In this process, CALB was more effective in that Pseudomonas cepacia lipase (PSL) showed opposite selectivity. Azetidine ring is an important structure because it is present in many compounds of pharmaceutical interest; however, its manipulation must be done very carefully owing to the reactivity of these heterocycles of small size. An interesting application of the use of biocatalytic processes is the resolution of azetidine esters (Scheme 7.11). The procedure to choose for the resolution of these compounds is the enzymatic ammonolysis of the corresponding N-substituted azetidines [26]. A derivative of the anti-inﬂammatory ibuprofen, the corresponding (S)-2-chloroethyl ester, has been obtained with 96% ee by CALB-catalyzed ammonolysis of the

O

O OMe

OMe

O

NH2

CALB

NR

NH3 sat. tBuOH 35 ºC

R = Benzyl, p-methoxybenzyl, allyl

+ NR (R)-Ester > 99% ee

Scheme 7.11 Enzymatic resolution of azetidinoesters.

NR (S)-Amide

7.4 Aminolysis and Ammonolysis of Esters Cl

O

Cl O

CALB NH3

O

NH2 +

O

O

t

BuOH 48 h, rt 56% conversion

(R )-Amide (S)-Ester 96% ee

Scheme 7.12 Resolution of an ibuprofen ester derivative by enzymatic ammonolysis.

corresponding racemic ester (Scheme 7.12) [27]. Reaction was performed in tert-butyl alcohol as solvent and bubbling ammonia through the solution. The resolution of racemic ethyl 2-chloropropionate with aliphatic and aromatic amines using Candida cylindracea lipase (CCL) [28] was one of the ﬁrst examples that showed the possibilities of this kind of processes for the resolution of racemic esters or the preparation of chiral amides in benign conditions. Normally, in these enzymatic aminolysis reactions the enzyme is selective toward the (S)-isomer of the ester. Recently, the resolution of this ester has been carried out through a dynamic kinetic resolution (DKR) via aminolysis catalyzed by encapsulated CCL in the presence of triphenylphosphonium chloride immobilized on Merriﬁeld resin (Scheme 7.13). This process has allowed the preparation of (S)-amides with high isolated yields and good enantiomeric excesses [29]. The strategy described here explains the different possibilities of enzymatic ammonolysis and aminolysis reaction for resolution of esters or preparation of enantiomerically pure amides, which are important synthons in organic chemistry. This methodology has been also applied for the synthesis of pyrrolidinol derivatives that can be prepared via enzymatic ammonolysis of a polyfunctional ester, such as ethyl ()-4-chloro-3-hydroxybutanoate [30]. In addition, it is possible in the resolution of chiral axe instead of a stereogenic carbon atom. An interesting enzymatic aminolysis of this class of reaction has been recently reported by Aoyagi et al. [31]. The side chain of binaphthyl moiety plays an important role in the enantiodiscrimination of the process (Scheme 7.14). Here, we have selected a few representative examples of the enzymatic resolution of esters by aminolysis or ammonolysis reactions. On the other hand, the enzymatic acylation of racemic amines is also of great utility for the preparation of optically pure O

O OEt Cl

CCL RNH2

+

PPh3 Cl

–

NHR Cl R=p-OMe-Ph 92% yield 86% ee

O OEt Cl

Scheme 7.13 Dynamic kinetic resolution of 2-Chloropropionate by enzymatic aminolysis.

j179

j 7 Aminolysis and Ammonolysis of Carboxylic Acid Derivatives

180

O

CN

+ H2N OEt

Lipase

O

TBME

N H

O

+

CN

OEt

(R)-Amide

(S)-Ester

Scheme 7.14 Resolution of a binaphthyl ester by an ammonolysis reaction.

amines and amides. Next, we have summarized some examples related to the utility of these enzymatic processes. 7.4.3 Resolution of Amines, Diamines, and Aminoalcohols

The enzymatic KR between racemic amines and nonactivated esters using a lipase as biocatalyst is shown in Scheme 7.15. In the same manner as in the transesteriﬁcation of secondary alcohols, this process ﬁts Kazlauskas rule [32], where normally if the large group (L) has larger priority than medium group (M), the (R)-amide is obtained. In general, major size differences between both groups result in better enantioselectivities (E). Scheme 7.16 is an example of a general strategy to obtain enantiopure amines, which can be used for the preparation of a great variety of chiral amines and amides. In many cases, for the isolation of the amine, it is convenient to carry out the synthesis through formation of the corresponding salt, or by protection of the amino group, depending on the stability of the starting amine. Normally ethyl acetate is used for the acylation of primary amines, in many cases, as acyl donor and solvent. Other acylating agents such as alkyl methoxy acetates are O

NH2 L

M

O +

1

Lipase

H HN L M

2

R

OR

R1

H2N H +

L

M

Scheme 7.15 Schematic representation of Kazlauskas rule in the enzymatic KR of an amine. O NH2 1

R

O 2

+

R

3

R

NH2

Lipase 4

OR

1

R

2

R

(S)-Amine

+

HN

R3

R1 R2 (R)-Amide Acid liquid extraction

(S)-Amine

Basic liquid extraction

(S)-Amine-H+

Scheme 7.16 Enzymatic resolution of amines.

(R)-Amide

7.4 Aminolysis and Ammonolysis of Esters Ph ( )n

Ph

Ph

NH2

NH2

( )n NH2

n = 1, 2

n = 1, 2

Scheme 7.17 Kinetic resolution of cis and trans-2-phenylcycloalkanamines.

also of utility; however, vinyl esters, the best reagents for the resolution of alcohols, are not adequate reagents for the resolution of primary amines owing to their high reactivity. In some cases, the enantiomeric excess can be increased by using ethyl esters of long chain fatty acids. For instance, it has been described that ethyl decanoate is a better acyl agent than other esters of shorter chain for the resolution of sec-butylamine [33]. In recent years, a great variety of primary chiral amines have been obtained in enantiomerically pure form through this methodology. A representative example is the KR of some 2-phenylcycloalkanamines that has been performed by means of aminolysis reactions catalyzed by lipases (Scheme 7.17) [34]. Kazlauskas rule has been followed in all cases. The size of the cycle and the stereochemistry of the chiral centers of the amines had a strong inﬂuence on both the enantiomeric ratio and the reaction rate of these aminolysis processes. CALB showed excellent enantioselectivities toward trans-2-phenylcyclohexanamine in a variety of reaction conditions (E > 150), but the reaction was markedly slower and occurred with very poor enantioselectivity with the cis-isomer, whereas Candida antarctica lipase A (CALA) was the best catalyst for the acylation of cis-2-phenylcyclohexanamine (E ¼ 34) and trans-2-phenylcyclopropanamine (E ¼ 7). Resolution of both cis- and trans-2-phenylcyclopentanamine was efﬁciently catalyzed by CALB obtaining all stereoisomers with high enantiomeric excess. The enantiodiscrimination of the biocatalysts is not only restricted to carbon stereogenic centers, but also other chirality elements such as chirality axe or atropoisomerism have been efﬁciently resolved through lipase-catalyzed acylation of the amino group. Optically active 1,10 -binaphthylamine derivatives are very useful for chiral ligands in asymmetric synthesis. In Scheme 7.18 the KR of three binaphthylamines via lipase-catalyzed amidation is shown. In this case, two immobilized lipases from Pseudomonas aeruginosa were most efﬁcient than CALB [35]. The enantioselectivity depends on the alkyl chain length between the binaphthyl ring and the amino group.

Acyl donor Lipase ( )n

NH2

n = 0,1, 2

i

Pr2O

O ( )n

N H

+ R

n = 0, (S)-enantiomer n = 1, (R)-enantiomer n = 2, (R)-enantiomer

Scheme 7.18 Enzymatic resolution of binapthylamines.

( )n

NH2

n = 0, (R)-enantiomer n = 1, (S)-enantiomer n = 2, (S)-enantiomer

j181

j 7 Aminolysis and Ammonolysis of Carboxylic Acid Derivatives

182

1

R

2

R

+ Me

HN

Na2CO3 Toluene 90 ºC

Me

O

R

O

CALB catalyst (4 mol%)

Me

O

NH2

R1

Me

R

2

R

ee = 93–99% 78–95% yield

O

H

O

R

R

R

R

H

Ru R R Ru OC CO CO CO Ruthenium catalyst (R= p-MeO-C6H4)

Scheme 7.19 Dynamic kinetic resolution of primary amines.

A KR is a useful method to obtain enantiomerically pure compounds, but suffers from the drawback that the maximum yield is only 50% of the starting material. This limitation can be overcome via a DKR process. This procedure can theoretically lead to a single product enantiomer with 100% yield. This strategy has been widely used for the resolution of alcohols by enzymatic hydrolysis or transesteriﬁcation processes, the acylation of alcohols being the most explored process [36]. This methodology has also been used to obtain enantiomerically pure amines. Recently, it has been reported a versatile DKR of several primary amines using a ruthenium catalyst for the racemization process, CALB, isopropyl acetate, and sodium carbonate (Scheme 7.19) [37]. An efﬁcient DKR process takes place obtaining the corresponding amides with high yields and enantiomeric excesses; however, still a high temperature (90 C) is needed. The power of biocatalysts for the production of chiral compounds can be elevated to a second stage when proper use of the process leads to a larger number of different enantioenriched products from the same reaction. Some efforts have been made in this direction because the enzyme can be selective to either nucleophile or acyl donor. The elegancy of the reaction is increased when the process is carried out for the resolution of both. Scheme 7.20 depicts this possibility [38]. For this reaction, CALB catalyzes the amidation between a racemic b-hydroxyester and racemic amines, leading to the corresponding amide with very high enantiomeric and diastereomeric excesses. Besides, the remaining ester and amine are recovered from the reaction media, also showing good enantiomeric excesses. By this method, three enantioenriched interesting compounds are obtained from an easy one-step reaction. Another philosophy would be the one-pot resolution of two different nucleophiles, alcohol and amine [39]. An acylated racemic alcohol reacts with a racemic amine in Me

OH O +

Cl

OEt

H2N

R

CALB

Me

OH O Cl

N H

+ Cl R

R = Phenyl, furyl, pentyl

Me

OH O + OEt

H2N

R

Scheme 7.20 Double resolution of both racemic esters and amines.

7.4 Aminolysis and Ammonolysis of Esters O Me

O 1

R

O +

2

R

R

Me OH O CALB + + 1 2 2 1,4-Dioxane R1 R R R

NH2 3

4

R

O O

O

Me

NH2

O HN 3

NH2 Me

Me

Me +

R

NH2 3

R

R4

O O

Me Me

NH2

4

R

O Me

j183

NH2

Me OMe

NH2

Me

F3C

Scheme 7.21 One-pot resolution of alcohols and amines.

the presence of CALB to yield four separable enantioenriched compounds. The (R)-alcohol acts as the leaving group in the acylation of the (R)-amine. The remaining (S)-ester and (S)-amine are also recovered from the reaction media (Scheme 7.21). Apart from the synthetic utility of the reaction, the authors use it for studying the effect of the leaving group in the resolution of the nucleophile. In the case of bifunctional compounds, two KRs over the two functional groups are very practical to achieve enantiopure compounds even if in the two steps the enantioselectivity obtained is moderated. A sequential biocatalytic resolution by one-pot double aminolysis has been reported for the resolution of cyclic 1,2-diamines obtaining the ﬁnal diacylated products from trans-cyclohexane-1,2-diamine and transcyclopentane-1,2-diamine in enantiopure forms [40]. The reaction is being carried out with dimethyl malonate using CALB as biocatalyst. The formation of the (R,R)bisamidoester involves two biocatalytic steps and the enzyme shows the same stereochemical preference toward the (R,R)-enantiomer of the substrate in both steps. A detailed study of the enantiomeric ratio for each step allowed us to suggest an interesting structural effect. The enantioselectivity of the second step is always higher than the ﬁrst one, in good agreement with Kazlauskas rule, because the monoacylated compound has a bigger steric difference in the substituents of the stereocenter than the free diamine. The mechanism for this sequential KR is shown in Scheme 7.22 for the case of trans-cyclohexane-1,2-diamine. The chemoenzymatic synthesis of the analgesic U-()-50,488 [41] and new C2-symmetric bisaminoamide ligands derived from N,N-disubstituted transcyclohexane,1,2-diamine [41] has been possible by a CALB-catalyzed resolution using ethyl acetate as solvent and acyl donor [42]. Aminoalcohols are an important class of compounds in medicinal chemistry because many drugs contain this structure. For their resolution, there are two possibilities: acylation of amino function or an enzymatic transesteriﬁcation with vinyl esters through the hydroxyl group. However, the amino or hydroxyl group must be protected, because if the starting material is the free aminoalcohol, the O- and N-acylation can take place, and in addition, there are migrations obtaining

NH2

j 7 Aminolysis and Ammonolysis of Carboxylic Acid Derivatives

184

Ser O NH2 Ser O NH2 NH2

O

OMe

Ser OH

O

OMe

O

O

O Ser OH

O NH OMe NH OMe

NH OMe (E2)

O O (R,R)-trans

O O (R,R)-trans

+

(E1)

NH2

(±)-trans NH2 (S,S)-trans

Scheme 7.22 Mechanism of the sequential kinetic resolution of trans-cyclohexane-1,2-diamine.

low enantioselectivities. For this reason, to achieve a good bioresolution with aminoalcohols, it is necessary to protect the amino or hydroxyl group. Normally, it is easier for the chemical acylation or alkoxycarbonylation of the amino group, and several examples have appeared recently in the literature about the enzymatic resolution of 1,2 and 1,3-aminoalcohols through an enzymatic transesteriﬁcation of the corresponding amides or carbamates as starting material derivatives [43]. Although the enzymatic resolution of 1,2-aminoalcohols has been exhaustively investigated, the bioresolution of 1,3-aminoalcohols has been scarcely reported, especially through enzymatic aminolysis processes. Recently, the enzymatic resolution of 3-amino-3-phenylpropan-1-ol derivatives has been studied by an enzymatic acylation process (Scheme 7.23). CALA has been identiﬁed as the best biocatalyst for the N-acylation reaction of 3-amino-3-phenyl-1-tert-butyldimethylsilyloxy-propan1-ol using ethyl methoxyacetate as acylating agent and tert-butyl methyl ether (TBME) as solvent [44]. This is not a surprising result as CALA has been identiﬁed as an ideal biocatalyst for the resolution of sterically hindered compounds [45]. The previous enzymatic study has allowed us to obtain a valuable intermediate for the production of (S)-Dapoxetine that has been synthesized in good overall yield and high optical purity. 7.4.4 Desymmetrization of Diesters

This methodology avoids the inherent 50% yield limit of KR and the difﬁcult separations often encountered in the resolution of racemates. The potential of enzymes, especially lipases, to catalyze the aminolysis and ammonolysis of prochiral O NH2

Me O Si tBu + MeO Me

O

CALA OEt TBME, 30 ºC

HN

OMe Me + O Si tBu Me

Scheme 7.23 Enzymatic resolution of an O-protected 1,3-aminoalcohol.

NH2

Me O Si tBu Me

7.5 Kinetic Resolution of Secondary Amines R1

MeO2C

R2NH2 CALB 1,4-Dioxane 30 ºC

CO2Me

1

1

R

2

R = OH, NHBn, OMe, OAc, Ph, p-F-C6H4

2

MeO2C

R = H, Bu, Bn

CONHR

76–99% ee

Scheme 7.24 Desymmetrization of prochiral glutarates.

substrates has been scarcely studied and only the desymmetrization of prochiral glutarates has been reported [46]. CALB is again the only biocatalyst that yields good results and reacts with the pro-(R)-ester group to yield the monoamidoester of (S) conﬁguration with very high yields and enantiomeric excesses (Scheme 7.24). In some cases, these products are excellent starting material for the synthesis of aminoacids of physiological interest [47].

7.5 Kinetic Resolution of Secondary Amines

The structure of the amine has a great inﬂuence in the lipase-catalyzed aminolysis reaction. Although primary amines can be efﬁciently resolved, few examples have been reported about the acylation or resolution of secondary amines. On the other hand, acyclic secondary amines are not good substrates for this enzymatic reaction. However, cyclic secondary amines are structurally better than acyclic compounds for their resolution, and several examples of resolution of pyrrolidine and piperidine derivatives have been described. Kanerva et al. reported the acylation of pipecolic acid derivatives catalyzed by CALA [48]. As explained above, this enzyme seems to have a larger pocket than CALB in the active site and accepts these bulkier substrates [45]. This reaction has also been applied to pyrrolidine ring and in both cases the enzyme was very efﬁcient, catalyzing the acylation of the secondary amino group with very high enantioselectivity (Scheme 7.25). Enzymatic alkoxycarbonylation gives better results by enzymatic resolution of 1-methyl tetrahydroisoquinoline than the corresponding acylation [49]. The best results are obtained with a 3-methoxyphenyl and allyl mixed carbonate as acylating reagent and Candida rugosa lipase (CRL) as biocatalyst. The best solvent is toluene and the (S)-amine and the (R)-carbamate were obtained with high yields and enantiomeric excesses (Scheme 7.26). Recently, Page and coworkers have reported an efﬁcient process for the DKR of this secondary amine using a novel iridium-based amine racemization catalyst that has allowed the preparation of the corresponding

O N H

OMe + O

R

O

CF3

CALA Solvent

OMe

N O

R

Scheme 7.25 Enzymatic resolution of pipecolic methyl esters.

O

+

N H

OMe O

j185

j 7 Aminolysis and Ammonolysis of Carboxylic Acid Derivatives

186

OMe CRL Toluene (0.05% H2O) 30 ºC, 8 h

O

NH + O

O

N

O

+

O (R)-Carbamate 98% ee 47% isolated yield

NH

(S)-Amine 99% ee 46% isolated yield

OMe CRL, [IrCp*I2]2

O

NH + O

O

N

Toluene 40 ºC, 23 h 90% conversion

O

O 96% ee 82% isolated yield

Scheme 7.26 Kinetic resolution and DKR of 1-methyl tetrahydroisoquinoline using Candida rugosa lipase.

R2

3

R

* * N

2

R3

R

O

1

R + N H

R4O

CALA or CALB O

TBME

2

+ O

R

3

R O

*

* N H

1

R

Scheme 7.27 Enzymatic preparation of enantiomerically pure indolines.

carbamate in 82% isolated yield and 96% ee using 3-methoxyphenyl propyl carbonate as alkoxycarbonylating agent [50]. An efﬁcient chemoenzymatic route for the synthesis of optically active substituted indolines has been recently developed (Scheme 7.27), and also the alkoxycarbonylation process is more efﬁcient than the acylation reaction. Different lipases have been tested in the alkoxycarbonylation of these secondary amines, CALA being found to be the best biocatalyst for 2-substituted-indolines, and CALB for 3-methylindoline. The combination of lipases with a variety of allyl carbonates and TBME as solvent has allowed the isolation of the carbamate and amine derivatives in a high level of enantiopurity [51].

7.6 Toward the Synthesis of b-Aminoacid Derivatives

Preparation of optically active b-aminoesters, b-aminonitriles, and b-aminocarboxamides are of special relevance for the synthesis of enantiomerically pure b-aminoacids compounds of special relevance in several areas of medicinal chemistry. The resolution of b-aminoesters can be carried out by acylation of the amino groups or by other biocatalytic reactions of the ester groups, such as hydrolysis, transesteriﬁcation, or aminolysis. The resolution of ethyl ()-3-aminobutyrate

7.6 Toward the Synthesis of b-Aminoacid Derivatives

j187

O NH2 O Me

O

CALB Solvent E = 74–88

+ Me

OEt

OEt

Me

NH O

NH2 O

+ Me

OEt

OEt CbzCl Na2CO3 H2O O CbzHN Me

NH O

O

+ Me

OEt

OEt NH2 O Me

RNH2

+ OEt

R = H, Bn

CbzHN

1) CALB, solvent 2) CbzCl, Na2CO3, H2O E = 4–16

O

CbzHN

Me

NHR

Me

Scheme 7.28 Enzymatic resolution of b-aminoesters.

is more efﬁcient by acylation of the amino group than by the aminolysis of the ester (Scheme 7.28) [52]. Other important derivatives for the preparation of b-aminoacids are the corresponding b-aminonitriles. Lipase-catalyzed N-acylations of racemic cis-2aminocyclopentane and cyclohexane carbonitriles with 2,2,2-triﬂuoroethyl butanoate have been successfully carried out in organic solvents and ionic liquids [53], PSL yielding better results than CALB (Scheme 7.29). Other derivatives to obtain b-aminoacids are the corresponding carboxamides. Thus, for the preparation of all enantiomers of cis and trans–2-aminocyclopentaneand cyclohexanecarboxamides, the best results obtained are using this acyl donor and CALB [54]. An unexpected change in enantiopreference accompanied by low enantioselectivity was observed when PSL (cis-cyclohexane substrate) or CALA (cis-cyclopentane and cyclohexane substrates) replaced CALB (Scheme 7.30). Recently, a very interesting preparation of b-peptides has been carried out by Kanerva and coworkers through a two-step lipase-catalyzed reactions in which N-acetylated b-amino esters were ﬁrst activated as 2,2,2-triﬂuoroethyl esters with CALB [55]. The activated esters were then used to react with a b-aminoester in the presence of CALA in dry diethylether or diisopropylether (Scheme 7.31). In this peptide synthesis, the aminoester was used in excess and the unreacted counterpart was easily separated and later recycled.

CN ( )n

O

Lipase

+ NH2

Pr

OR

Solvent

( )n

CN O N H

CN + ( )n Pr

(±)-cis, n = 1, 2

Scheme 7.29 Enzymatic resolution of b-aminonitriles.

O

+

NH2

OEt

j 7 Aminolysis and Ammonolysis of Carboxylic Acid Derivatives

188

O

O NH2

( )n

O

Lipase

+ Pr

O

CF3

O NH2

( )n

+

Solvent

NH2

( )n

Pr

HN

NH2 NH2

O

(±)-trans, n = 1, 2

(1R,2R) n = 1 (1R,2R) n = 2

(1S,2S) n = 1 (1S,2S) n = 2

Scheme 7.30 Enzymatic resolution of b-aminocarboxamides.

1

O Me

R N H

O

O OEt

+

1

O

R

O

CALB

Pr

O

CF3

Me

N H

2

R

1

R

O

CF3

R2

O CALA

3

H2N

OR 2

R

1

O Me

N H

3

O

R

2

R

O

R N H

OR5 4

R

Scheme 7.31 Lipases in b-dipeptide synthesis in organic solvents.

7.7 Summary and Outlook

Hydrolytic enzymes are the most used biocatalysts, and the application of the biocatalysis in organic synthesis is currently a well-established methodology for the chemo- and regioselective modiﬁcation of nonchiral compounds, resolution of racemates, and desymmetrization of prochiral substrates. In this chapter, we have shown how hydrolases, especially lipases, catalyze enzymatic aminolysis and ammonolysis of carboxylic acids and esters in different solvent systems, providing a potential application for the preparation of amines and amides, compounds which play an important role in the production of ﬁne chemicals and pharmaceuticals. Presently, it is clear that for the resolution of primary amines CALB is the best biocatalyst; however, CALA results are more appropriate for the resolution of secondary amines. In addition, the combination of genetic engineering and molecular modeling techniques are playing a major role in the development of enzymes that will show better results in the future than those currently achieved, and will offer new pathways and possibilities for the synthesis and resolution of interesting manufactures for the industrial sector.

References

References 1 Ghanem, A. and Aboul-Enein, H.Y. (2004) Tetrahedron: Asymmetry, 15, 3331–3351. 2 Garcıa-Urdiales, E., Alfonso, I. and Gotor, V. (2005) Chemical Reviews, 105, 313–354. 3 Torres-Gavilan, A., Castillo, E. and LópezMunguıa, A. (2006) Journal of Molecular Catalysis B: Enzymatic, 41, 136–140. 4 (a) Gotor, V. (1999) Bioorganic and Medicinal Chemistry, 7, 2189–2197. (b) van Rantwijk, F. and Sheldon, R.A. (2004) Tetrahedron, 60, 501–519. (c) Alfonso, I. and Gotor, V. (2004) Chemical Society Reviews, 33, 201–209. (d) GotorFernandez, V. and Gotor, V. (2006) Current Organic Chemistry, 10, 1125–1143. 5 Gotor-Fernandez, V., Rebolledo, F. and Gotor, V. (2007) Preparation of chiral pharmaceuticals through enzymatic acylation of alcohols and amines, in Biocatalysis in the Pharmaceutical and Biotechnology Industry, (ed. R.M. Patel), Dekker, Taylor and Francis, New York, Chapter 7, pp. 203–248. 6 Hayes, K.S. (2001) Applied Catalysis A-General, 221, 187–195. 7 Gotor Fernandez, V., Busto, E. and Gotor, V. (2006) Advanced Synthesis and Catalysis, 348, 797–812. 8 (a) Astorga, C., Rebolledo, F. and Gotor, V. (1993) Synthesis, 287–289. (b) Hacking, M.A.P.J., Akkus, H., van Rantwijk, F. and Sheldon, R.A. (2000) Biotechnology and Bioengineering, 68, 84–91. 9 (a) Bornscheuer, U.T. and Kazlauskas, R.J. (2006) Hydrolases in Organic Synthesis, 2nd edn, Wiley-VCH Verlag GmbH. (b) Faber, K. (2004) Biotransformations in Organic Chemistry, 5th edn, Springer-Verlag, Heildeberg. (c) Bommarius, A.S. and Riebel, B.R. (2004) Biocatalysis, Wiley-VCH Verlag GmbH, Weinheim. 10 Slotema, W.F., Sandoval, G., Guieysse, D., Straathof, A.J.L. and Marty, A. (2003)

11

12

13

14

15

16

17

18

19 20

21

22

Biotechnology and Bioengineering, 82, 664–669. Madeira Rau, R., Rantwijk, F.V., Seddon, K.R. and Sheldon, R.A. (2000) Organic Letters, 2, 4189–4191. (a) Irimescu, R. and Kato, K. (2004) Journal of Molecular Catalysis B: Enzymatic, 30, 189–194. (b) Irimescu, R. and Kato, K. (2004) Tetrahedron Letters, 45, 523–525. Prasad, A.K., Husain, M., Singh, B.K., Gupta, R.K., Manchanda, V.K., Olsen, C.E. and Parmar, V.S. (2005) Tetrahedron Letters, 46, 4511–4514. Ulijn, R.U., Baragaña, B., Halling, P.J. and Flitsch, S.L. (2002) Journal of the American Chemical Society, 124, 10988–10989. Furutani, T., Furui, M., Ooshima, H. and Kato, J. (1996) Enzyme and Microbial Technology, 19, 578–584. Levinson, W.E., Kuo, T.M. and Kurtzman, C.P. (2005) Enzyme and Microbial Technology, 37, 126–130. (a) Fernandez-Perez, M. and Otero, C. (2001) Enzyme and Microbial Technology, 28, 527–536. (b) Fernandez-Perez, M. and Otero, C. (2003) Enzyme and Microbial Technology, 33, 650–660. Tufvesson, P., Annerling, A., Hatti-Kaul, R. and Adlecreutz, D. (2007) Biotechnology and Bioengineering, 97, 447–453. Liljeblad, A. and Kanerva, L.T. (2006) Tetrahedron, 62, 5831–5854. De Zoote, M.C., Kock van-Dalen, A.C., van Rantwijk, F. and Sheldon, R.A. (1996) Journal of Molecular Catalysis B: Enzymatic, 2, 141–145. Torre, O., Gotor-Fernandez, V., Alfonso, I., Garcıa-Alles, L.F. and Gotor, V. (2005) Advanced Synthesis and Catalysis, 347, 1007–1014. Gill, I. and Patel, R. (2006) Bioorganic and Medicinal Chemistry Letters, 16, 705–709.

j189

j 7 Aminolysis and Ammonolysis of Carboxylic Acid Derivatives

190

23 Quirós, M., Astorga, C., Rebolledo, F. and Gotor, V. (1995) Tetrahedron, 51, 7715–7720. 24 Quan, J., Wang, N., Cai, X.-Q., Wu, Q. and Lin, X.-F. (2007) Journal of Molecular Catalysis B: Enzymatic, 44, 1–7. 25 Garcıa, M.J., Rebolledo, F. and Gotor, V. (1993) Tetrahedron: Asymmetry, 4, 2199–2210. 26 Starmans, W.A.J., Doppen, R.G., Thijs, L. and Zwanemburg, B. (1998) Tetrahedron: Asymmetry, 9, 429–435. 27 de Zoete, M.C., Kock-van Dalen, A.C., van Rantwijk, F. and Sheldon, R.A. (1993) Journal of the Chemical Society, Chemical Communications, 1831–1832. 28 Gotor, V., Rebolledo, F. and Brieva, R. (1988) Tetrahedron Letters, 29, 6973–6974. 29 BadjicJ.D., Kadnikova, E.N. and Kostic, N.M. (2001) Organic Letters, 3, 2025–2028. 30 Garcıa-Urdiales, E., Rebolledo, F. and Gotor, V. (1999) Tetrahedron: Asymmetry, 10, 721–726. 31 Aoyagi, N., Kawauchi, S. and Izumi, T. (2004) Tetrahedron Letters, 45, 5189–5192. 32 Kazlauskas, R.J., Weissﬂoch, A.N.E., Rappaport, A.T. and Cuccia, L.A. (1991) Journal of Organic Chemistry, 56, 2656–2665. 33 Goswami, A., Guo, Z., Parker, W.L. and Patel, R.N. (2005) Tetrahedron: Asymmetry, 16, 1715–1719. 34 (a) Gonzalez-Sabin, J., Gotor, V. and Rebolledo, F. (2005) Tetrahedron: Asymmetry, 15, 481–488. (b) GonzalezSabin, J., Gotor, V. and Rebolledo, F. (2005) Tetrahedron: Asymmetry, 16, 3070–3076. 35 Aoyagi, N. and Izumi, T. (2002) Tetrahedron Letters, 43, 5529–5531. 36 Pamies, O. and B€ackvall, J.E. (2003) Chemical Reviews, 103, 3247–3262. 37 Paetzold, J. and B€ackvall, J.E. (2005) Journal of the American Chemical Society, 127, 17620–17621. 38 Sanchez, V.M., Rebolledo, F. and Gotor, V. (1999) Journal of Organic Chemistry, 64, 1464–1470.

39 Garcıa-Urdiales, E., Rebolledo, F. and Gotor, V. (2000) Tetrahedron: Asymmetry, 11, 1459–1463. 40 (a) Alfonso, I., Astorga, C., Rebolledo, F. and Gotor, V. (1996) Chemical Communications, 2471–2472. (b) Luna, A., Alfonso, I. and Gotor, V. (2002) Organic Letters, 4, 3627–3629. 41 Gonzalez-Sabin, J., Gotor, V. and Rebolledo, F. (2004) Chemistry - A European Journal, 10, 5788–5794. 42 Gonzalez-Sabin, J., Gotor, V. and Rebolledo, F. (2006) Tetrahedron: Asymmetry, 17, 449–454. 43 (a) Maestro, A., Astorga, C. and Gotor, V. (1997) Tetrahedron: Asymmetry, 8, 3153–3159 (b) López-Garcıa, M., Alfonso, I. and Gotor, V. (2004) Chemistry- A European Journal, 10, 3006–3014. 44 Torre, O., Gotor-Fernandez, V. and Gotor, V. (2006) Tetrahedron: Asymmetry, 16, 860–866. 45 Domınguez de Marıa, P., CarboniOerlemans, C., Tuin, B., Bargeman, G., Van de Meer, A. and Van Gemert, R. (2005) Journal of Molecular Catalysis B: Enzymatic, 37, 36–46. 46 (a) Puertas, S., Rebolledo, F. and Gotor, V. (1996) Journal of Organic Chemistry, 61, 6024–6027. (b) Jacobsen, E.E., Hoff, B.H., Moen, A.R. and Anthonsen, T. (2003) Journal of Molecular Catalysis B: Enzymatic, 21, 55–58. (c) López-Garcıa, M., Alfonso, I. and Gotor, V. (2003) Tetrahedron: Asymmetry, 14, 603–609. 47 López-Garcıa, M., Alfonso, I. and Gotor, V. (2003) Journal of Organic Chemistry, 68, 648–665. 48 Liljeblad, A., Lindborg, J., Kanerva, A., Katajisto, J. and Kanerva, L.T. (2002) Tetrahedron Letters, 43, 2471–2474. 49 Breen, G.F. (2004) Tetrahedron: Asymmetry, 15, 1427–1430. 50 Stirling, M., Blacker, J. and Page, M.I. (2007) Tetrahedron Letters, 48, 1247–1250. 51 Gotor-Fernandez, V., Fernandez-Torres, P. and Gotor, V. (2006) Tetrahedron: Asymmetry, 17, 2558–2564.

References 52 Sanchez, V.M., Rebolledo, F. and Gotor, V. (1997) Tetrahedron: Asymmetry, 8, 37–40. 53 Fitz, M., Lundell, K., Lindros, M., F€ ul€op, F. and Kanerva, L.T. (2005) Tetrahedron: Asymmetry, 16, 3690–3697.

54 Fitz, M., Lundell, K., Lindros, M., F€ ul€op, F. and Kanerva, L.T. (2006) Tetrahedron: Asymmetry, 17, 1129–1134. 55 Li, X.-G. and Kanerva, L.T. (2006) Organic Letters, 8, 5593–5596.

j191

j193

8 Enzymatic Reduction Reaction Kaoru Nakamura and Tomoko Matsuda

8.1 Introduction

Reduction is one of the most important, fundamental, and practical reactions – for example, reduction of ketones affords chiral compounds, which can be transformed into suitable forms with various functionalities to synthesize industrially important chemicals such as pharmaceuticals, agrochemicals, and natural products. The catalysts for the asymmetric reduction can be classiﬁed into two categories: chemical and biological. Both have their own peculiarities, and development of both to enable the appropriate selection of the catalysts for particular purposes is necessary to promote green chemistry. For example, biocatalysts have the advantages of being natural, having high chemo-, regio-, and enantioselectivity, and being active under mild reaction conditions. Here, new methodology for improving reactivity and selectivity of enzymatic reduction will be given. Synthetic applications of enzymatic reductions are also discussed.

8.2 Hydrogen Source for Coenzyme Regeneration

Enzymes that perform reduction of carbonyl groups usually require a coenzyme from which a hydride is transferred to the carbonyl carbon. Since reduction of the substrate is concomitant with oxidation of the coenzyme, and the coenzyme is too expensive as a throwaway reagent, it is necessary to recycle and reuse the oxidized form of the coenzyme [1]. The oxidized form of the coenzymes has to be transformed to the reduced form for the next cycle of reduction of substrate. Hydrogen sources are necessary to perform this reduction reaction. For biocatalytic reduction, alcohols such as ethanol and 2-propanol, sugars such as glucose, glucose-6-phosphate, and glucose-6-sulfate, formic acid, and dihydrogen can be used. Some examples are shown in this section.

Asymmetric Organic Synthesis with Enzymes. Edited by Vicente Gotor, Ignacio Alfonso, and Eduardo Garcıa-Urdiales Copyright 2008 WILEY-VCH Verlag GmbH Co. KGaA, Weinheim ISBN: 978-3-527-31825-4

j 8 Enzymatic Reduction Reaction

194

Geotrichum candidum O

OH Ph

Ph NADH

Cofactor recycling

NAD+

O

OH

Figure 8.1 NADH recycling using alcohol as a hydrogen source for reduction [2].

8.2.1 Alcohol as a Hydrogen Source of Reduction

Alcohols such as ethanol, 2-propanol, and so on, have been widely used to recycle the coenzyme for the reduction catalyzed by alcohol dehydrogenase since the enzyme catalyzes both reduction and oxidation. Usually, an excess amount of the hydrogen source is used to push the equilibrium toward formation of product alcohols. There is an interesting example of the use of secondary alcohols in which the enantioselectivity and reaction yield was improved by recycling the coenzyme using secondary alcohols for the reduction of ketones with Geotrichum candidum Figure 8.1 [2]. Although the enantioselectivity of the reduction of acetophenone with resting cell of the microbe was low, the use of the dried cells as a biocatalyst and addition of a catalytic amount of NAD(P)þ and excess amount of secondary alcohols such as 2-propanol and cyclopentanol increased enantioselectivity and also chemical yield of the reduction; the (S)-alcohol was obtained in >99% ee with high yield. However, the addition of stoichiometric amount of NAD(P)H in the dried-cell system gave low enantioselectivity. 8.2.2 Sugar as a Hydrogen Source of Reduction

Glucose and glucose-6-phosphate have been widely used as a reducing energy. Recent example shows that a thermostable glucose-6-phosphate dehydrogenase (G6PDH) from Bacillus stearothermophillus was used for recycling NADPH at high temperature (55 C) for the reduction of 2-butanone by a thermostable alcohol dehydrogenase from Thermoanaerobacter brockii. Glucose-6-sulfate was used instead of glucose-6phosphate and the sulfate is three times more effective than the phosphate, the natural substrate for G6PDH (Figure 8.2) [3]. 8.2.3 Formate as a Hydrogen Source of Reduction

Formate is one of the most representative hydrogen sources for the biocatalytic reduction because CO2 formed by the oxidation of formate is released easily from the reaction system [4]. For example, for the reduction of aromatic ketones by the

8.2 Hydrogen Source for Coenzyme Regeneration

6-Sulfo-gluconolactone

NADPH

j195

2-Butanone

ADH G6PDH (Bacillus stearothermophillus) (Thermoanaerobacter brockii) NADP+

Glucose-6-sulfate

2-Butanol

Figure 8.2 Reduction of ketone with alcohol dehydrogenase from Thermoanaerobacter brockii using glucose-6-sulfate as a hydrogen source [3].

recombinant (S)-alcohol dehydrogenase from Rhodococcus erythropolis, formate dehydrogenase (FDH) from Candida boidinii was used for regeneration of NADH. By using hexane as an organic solvent in biphasic reaction medium, substrate concentrations of 10–200 mM can be used, and the resulting (S)-alcohols were formed with moderate to good conversion rates, and with up to >99% ee (Figure 8.3) [4]. Formate-FDH system was also applied in the reduction of 6-bromotetralone to (S)-6-bromotetralol, a potential pharmaceutical precursor, with the NADH-dependent ketone reductase from Trichosporon capitatum [4b]. A resin (XAD L-323) was used to bind the product (Figure 8.4). 8.2.4 Molecular Hydrogen as a Hydrogen Source of Reduction

Molecular hydrogen has been used for the recycling of coenzymes [5]. For example, the soluble hydrogenase I (H2 : NADPþ oxidoreductase, EC 1.18.99.1) from the marine hyperthermophilic strain of the archaeon Pyrococcus furiosus (PF H2ase I) has been used as a biocatalyst in the enzymatic production and regeneration of NADPH utilizing molecular hydrogen. Utilizing the thermophilic NADPH-dependent alcohol dehydrogenase from Thermoanaerobium species (ADH M) coupled to the PF H2ase I in situ NADPH-regenerating system, two prochiral model substrates, acetophenone and (2S)-hydroxy-1-phenyl-propanone, were quantitatively reduced

O

O H3C

H3C CO2

+

NADPH + H

Formate dehydrogenase (Candida boidini)

ADH (Rhodococcus erythropolis )

Hexane layer OH

OH

NADP+

HCO2H

Cl

Cl

H3C

H3C Aqueous layer 95% conversion ee > 99% Figure 8.3 Reduction of ketone with alcohol dehydrogenase from Rhodococcus erythropolis using formate as a hydrogen source [4a].

Cl

Cl

j 8 Enzymatic Reduction Reaction

196

O

XAD L-323 reductase (Trichosporon capitatum)

Br

OH (S) Br

NADH

NAD+

CO2

HCO2H FDH (Candida boidinii)

Figure 8.4 Reduction of 6-bromotetralone with reductase from Trichosporon capitatum using formate as a hydrogen source [4b].

to the corresponding (S)-alcohol and (1R,2S)-diol in which >99.5% ee and >98% de, respectively, with total turnover numbers (ttn: mol product/mol consumed cofactor NADPþ) of 100 and 160 could be reached (Figure 8.5) [5a]. Another example for the use of hydrogen as reductant is observed in the reduction of imine [5b]. New imine reductase activity has been discovered in the anaerobic bacterium Acetobacterium woodii by screening a dynamic combinatorial library of virtual imine substrates, using a biphasic water–tetradecane solvent system. A ruthenium complex [RuCl2(TPPTS)2]2 was used for regeneration of NADPþ to NADPH with hydrogen. Thus, 2-heptanone was reduced with alcohol dehydrogenase from Thermoanaerobacter brockii in the presence of the ruthenium complex, NADPþ, and hydrogen at 60 C to (S)-2-heptanol in 40 % ee. Turnover number was reported to be 18 (Figure 8.6) [5c]. 8.2.5 Light Energy as a Hydrogen Source of Reduction

Photochemical methods [6] have been developed to provide an environmentally friendly system that employs light energy to regenerate NAD(P)H, for example, by the use of a cyanobacterium, a photosynthetic biocatalyst. Using the biocatalysts, the O

OH

H3C

99.5% ee

H3C ADH (Thermoanaerobacter sp.) O

OH

H3C

H3C NADPH

OH

NADP+

H2 Hydrogenase (Pyrococcus furiosus ) Figure 8.5 Reduction of ketone with alcohol dehydrogenase from Thermoanaerobacter species using molecular hydrogen as a hydrogen source [5a].

OH

> 99.5% ee

8.2 Hydrogen Source for Coenzyme Regeneration

O NADPH + H + [RuCl2(TPPTS) 2]2 H2

TBADH

OH

(S) 40% ee

NADP +

Figure 8.6 Reduction of ketone with ruthenium complex and alcohol dehydrogenase using molecular hydrogen as a hydrogen source [5c].

reduction of acetophenone derivatives occurred more effectively under illumination than in the dark (Figure 8.7) [6b,c]. The light energy harvested by the cyanobacterium is converted into chemical energy in the form of NADPH through an electron transfer system, and, consequently, the chemical energy (NADPH) is used to reduce the substrate to chiral alcohol (96 to >99% ee). The light energy, which is usually utilized to reduce CO2 to synthesize organic compounds in the natural environment, was used to reduce the substrate in this case. When a photosynthetic organism is omitted, the addition of a photosensitizer is necessary. The methods use light energy to promote the transfer of an electron from a photosensitizer to NAD(P)þ via an electron transport reagent [6g]. Recently, carbon dioxide was reduced to formic acid with FDH from Saccharomyces cerevisiae in the presence of methylviologen (MV2þ) as a mediator, zinc tetrakis(4-methylpyridyl) porphyrin (ZnTMPyP) as a photosensitizer, and triethanolamine (TEOA) as a hydrogen source (Figure 8.8) [6h].

Light

Light-harvesting system

H2O O

Electron transfer system

NADPH

Cofactor+ recycling NADP

O2 ee 96 – > 99% H OH

X

X (S)

CO2

Calvin cycle

Organic compounds

Cyanobacteria Figure 8.7 Reduction of ketone with photosynthetic biocatalyst using light energy [6b,c].

j197

j 8 Enzymatic Reduction Reaction

198

Formate DH CO2

HCO2H

(HOCH2CH2)3N Light, ZnTMPyP MV2+

Figure 8.8 Reduction of carbon dioxide with formate dehydrogenase and porphyrin complex using light energy [6h].

8.2.6 Electric Power as a Hydrogen Source of Reduction

Electrochemical regeneration of NAD(P)H has long been recognized as a potentially powerful technology from the viewpoint of green chemistry, as it would not require a second enzyme and cosubstrate [1b]. However, the method is not effective because of the necessity of high overpotentials with direct reduction of the coenzyme, electrode fouling, dimerization of the coenzyme, and the fact that only the enzyme in the immediate vicinity of the electrode will be productive. Viologen-diaphorase (lipoamide dehydrogenase) were used for reduction of NAD(P)þ, where viologens were used as a mediator of an electron from electrode to diaphorase (Figure 8.9) [7a]. Organometallics such as rhodium complex were also used for electrochemical regeneration of NAD(P)H from electrode (Figure 8.10) [7b]. Surface-modiﬁed electrodes were used for prevention of high overpotentials with direct oxidation or reduction of the cofactor, electrode fouling, and dimerization of the cofactor [7c]. Membrane electrochemical reactors were designed. The regeneration of the cofactor NADH was ensured electrochemically, using a rhodium complex as electrochemical mediator. A semipermeable membrane (dialysis or ultraﬁltration) was integrated in the ﬁlter-press electrochemical reactor to conﬁne MV+ e–

MV2+

DI'

DIred

NAD+

MV+

DIox

DIox

NADH

Figure 8.9 Regeneration of NADH using diaphoras and electric power. DI: diaphorase [7a].

2+ N 2e–

N Rh

Cp

NADH

OH2 +

H+ N

N

NAD+

Rh Cp

H

Figure 8.10 Regeneration of NADH with rhodium complex using electric power [7b].

8.3 Methodology for Stereochemical Control

[Cp(Me)5Rh(bipy)Cl]+

Cyclohexanone

NADH

ADH

2e– [Cp(Me)5Rh(bipy)H]+

NAD+

Electrode

Cyclohexanol Membrane

Figure 8.11 Reduction of cyclohexanone with alcohol dehydrogenase and rhodium complex using electric power [7b].

the enzyme(s) as close as possible to the electrode surface. Cyclohexanone was reduced effectively to cyclohexanone using horse liver alcohol dehydrogenase (HLADH) (Figure 8.11) [7d,e]. ToincreasetheefﬁciencyofelectrochemicalreductionofNADPþ,apolyaminoanilinemodiﬁed electrode was prepared, and the reduction of NADPþ by ferredoxine NADPþ reductase was investigated. Although the immobilization of only the copolymerized viologen showed no activity, immobilization of both viologen and NADPþ on the electrode showedthe reduction wave at1.2 V versus SCE (saturated calomel electrode), corresponding to that of free NADPþ on the electrode [7].

8.3 Methodology for Stereochemical Control

Since stereoselectivities of biocatalytic reductions are not always satisfactory, modiﬁcation of biocatalysis are necessary for practical use. This section explains how to ﬁnd, prepare, and modify the suitable biocatalysts, how to recycle the coenzyme, and how to improve productivity and enantioselectivity of the reactions. 8.3.1 Screening of Microbes

Screening for a novel enzyme is a classical method and still one of the most powerful tools for ﬁnding the biocatalytic reduction system [8]. It is possible to discover a suitable biocatalyst by applying the latest screening and selection technologies, allowing rapid identiﬁcation of enzyme activities from diverse sources [8a]. Enzyme sources used for screening for asymmetric reductions in organic synthesis can be soil samples, commercial enzymes, culture sources, or a clone bank, and so on. Their origin can be microorganisms, animals, or plants. From these sources, enzymes that are regularly expressed and those that are not expressed in the original host can be tested to determine whether they are suitable for the transformation of certain substrates [8a]. For example, a screening of 416 strains (71 bacterial strains, 45 actinomycetes, 59 yeast, 60 basidiomycetes, 33 marine fungi, and 148 ﬁlamentous fungi) has been performed to look for microorganisms that display reductase activity in the absence of oxidase activity [8b]. A new microorganism, Diplogelasinospora grovesii IMI

j199

j 8 Enzymatic Reduction Reaction

200

Suitable microbe

Substrate

Product

(a)

Screening

OH

O Diplogelasinospora grovesii

> 98% ee

Reference

71 bacterial strains, 45 actinomycetes, 59 yeast, 60 basidiomycetes, 33 marine fungi , 148 filamentous fungi

[8b]

(S)

(b) O Cl

OH

Candida magnoliae

CO2Et

Cl

CO2Et

–1

90 gl , ee 96.6% (ee 99% after heat treatment) (c)

[8c]

450 bacteria

[8d]

OH

Klebsiella pneumoniae IFO 3319

O

400 yeast

(S)

CO2Et

CO2Et Yield 99%, de 99% , ee >99% (2 kg in 200-l fermentor )

(2R, 3S)

(d) O Cl

O

OMe

80 microorganisms (yeast, fungi, bacteria)

[8e]

N

Microbacterium campoquemadoensis (MB5614) OH Cl

O

OMe

(S) 500 mgl– ee > 95%

N

(e)

KRED 101

O

OH

CO2Et CO2Et NADPH

CO2Et

Keto-reductase-screening kit

Glucose dehydrogenase

[8f]

CO2Et

(3S,4R)

Figure 8.12 Screening for a biocatalyst [8].

171018, was isolated and showed very high activity and stereoselectivity in the reduction of cyclic ketones. For example, the bicycloketone were reduced in >98% ee (S) (Figure 8.12a). In another example, 400 yeasts were screened for the reduction of ethyl 4-chloro-3oxobutanoate, and Candida magnoliae was found to be the best one, as shown in Figure 8.12b [8c]. For the reduction of ethyl 2-methyl-3-oxobutanoate, Klebsiella pneumoniae IFO 3319 out of 450 bacterial strains was found to give the corresponding (2R,3S)-hydroxy ester with 99% de and >99% ee (Figure 8.12c) [8d]. Screening techniques have also been applied to drug synthesis. For example, a key intermediate in the synthesis of the antiasthma drug, Montelukast, was prepared from the

8.3 Methodology for Stereochemical Control

corresponding ketone by microbial transformation as shown in Figure 8.12d. The biotransforming organism Microbacterium campoquemadoensis (MB5614) was discovered as a result of an extensive screening program [8e]. Keto-reductase-screening kit is now available (10 enzyme for 2295 from Biocatalytics Inc.,) and this kit was used for screening of reduction of keto diester (Figure 8.12e). KRED101 afforded (3S,4R)hydroxydiester in 90% de with 100% ee [8f ]. There is also an example for the systematic screening of microbes against aromatic ketones [8g]. A collection of about 300 microbes was surveyed for the ability to generate chiral secondary alcohols by enantioselective reduction of a series of alkyl aryl ketones. Microbial cultures demonstrating utility in reducing model ketones were arrayed in multiwell plates and used to rapidly identify speciﬁc organisms capable of producing chiral alcohols used as intermediates in the synthesis of several drug candidates. Approximately 60 cultures were shown to selectively reduce various ketones providing both the R and S enantiomers of the corresponding alcohols in 92–99% ee with yields up to 95% at 1–4 g l1. An alternative approach to chiral alcohols based on selective microbial oxidation of racemic alcohols is also reported. This study provides a useful reference for generating chiral alcohols by selective microbial bioconversion.

8.3.2 Modification of Biocatalysts by Genetic Methods 8.3.2.1 Modified Yeast Recently, various genetic methods for screening, as well as classical methods, have been reported in the search for effective biocatalytic reduction. One of the most interesting examples is the use of glutathione S-transferase (GST)-fusion proteins to allow rapid identiﬁcation of synthetically useful biocatalysts [9]. A set of fusion proteins consisting of GST linked to the N-terminus of putative dehydrogenases produced by bakers yeast (S. cerevisiae) was screened for the reduction of various substrates. For example, ethyl 2-oxo-4-phenylbutyrate was reduced rapidly in the presence of NADH and NADPH by two dehydrogenases, Ypr1p and Gre2p, providing the (R)- and (S)-alcohol, respectively, with high stereoselectivities (Figure 8.13a) [9a]. The same enzymes were overexpressed in their native forms in Escherichia coli and growing cells of the engineered strains could also be used to carry out the reductions without the need for exogenous cofactor. A representative set of a- and b-keto esters was also tested as substrates (total 11) for each puriﬁed fusion protein (Figure 8.13b,c) [9b]. The stereoselectivities of -keto ester reductions depended both on the identity of the enzyme and the substrate structure, and some reductases yielded both L- and D-alcohols with high stereoselectivities. While a-keto esters were generally reduced with lower enantioselectivities, it was possible to identify pairs of yeast reductases that delivered both alcohol antipodes in optically pure form. These results demonstrate the power of genomic fusion protein libraries to identify appropriate biocatalysts rapidly and expedite process development.

j201

j 8 Enzymatic Reduction Reaction

202

OH (a) CO2Et

[9a]

(R)

Ypr1p ee 97% (GST-Ypr1p, NADP+, G6P, G6PDH)

O

ee 87% (E. coli cell overexpressing Ypr1p) CO2Et

OH Gre2p

CO2Et

Dehydrogenase from Saccharomyces cerevisiae

(S)

ee 90% (GST-Gre2p, NADP+, G6P, G6PDH) ee 91% ( E.coli cell overexpressing Gre2p) OH CO2Et

R1 (b)

R2

OH

L-anti

CO2Et

R1

O

R2

CO2Et

R1

[9b]

OH

R2 Dehydrogenase from Saccharomyces cerevisiae

L-syn

CO2Et

R1

OH

R2

CO2Et

R1 R2

D-syn

D-anti

OH (c)

O R

R

CO2Et

L

[9b]

CO2Et

Dehydrogenase from Saccharomyces cerevisiae

OH R

CO2Et

D

Figure 8.13 Identification of appropriate biocatalysts from fusion protein libraries from bakers yeast [9].

8.3.2.2 Overexpression Biocatalysts, discovered by screening, can be prepared in a large quantity by overexpression of the enzymes in transformed E. coli [10,11]. For example, the gene encoding (6R)-2,2,6-trimethyl-1-4-cyclohexanedione (levodione) reductase was cloned from the genomic DNA of the soil-isolated bacterium Corynebacterium aquaticum M-13 (Figure 8.14a) [10a]. The enzyme was sufﬁciently produced in recombinant E. coli cells and the enzyme puriﬁed from E. coli catalyzed stereo- and regioselective reduction of levodione. The enzyme was strongly activated by monovalent cations, such as Kþ, Naþ, and NH4. The chiral intermediate (S)-1-(20 -bromo-40 -ﬂuorophenyl) ethanol was prepared by the enantioselective microbial reduction of 2-bromo-4-ﬂuoroacetophenone [10b]. Organisms from genus Candida, Hansenula, Pichia, Rhodotorula , Saccharomyces, Sphingomonas, and bakers yeast reduced the ketone to the corresponding alcohol in

8.3 Methodology for Stereochemical Control a)

O

Levodione reductase from C. aquaticum (Overexpressed in E. coli)

O (R)

O

[10a]

HO O

b)

j203

ADH from Pichia methanolica (Overexpressed in E. coli)

H3C X

F

OH H 3C

Glucose, Glucose DH

(S) X

F

[10b]

2 kg scale

X = Br, (CH2)3CO2R c)

O Cl

CO2Et

ADH from Candida parapsilosis (Overexpressed in E. coli) 2-Propanol

OH Cl

CO2Et

(R) [10c] ee > 99% –1 36.6 gl

Figure 8.14 Overexpressed reductase in E. coli [10].

>90% yield and 99% ee. In an alternative approach, the enantioselective microbial reductions of methyl, ethyl, and tert-butyl 4-(20 -acetyl-50 -ﬂuorophenyl) butanoates were demonstrated using strains of Candida and Pichia. Reaction yields of 40–53% and enantiomeric excesses of 90–99% were obtained for the corresponding (S)-hydroxyesters. The reductase, which catalyzed the enantioselective reduction of keto esters, was puriﬁed to homogeneity from cell extracts of Pichia methanolica SC 13825. It was cloned and expressed in E. coli with recombinant cultures used for the enantioselective reduction of keto methyl ester to the corresponding (S)-hydroxymethyl ester. On a preparative scale, a reaction yield of 98% and an enantiomeric excess of 99% were obtained (Figure 8.14b). The synthesis of ethyl (R)-4-chloro-3-hydroxy-butanoate ((R)-ECHB) from ethyl 4-chloroacetoacetate (ECAA) was studied using whole recombinant cells of E. coli expressing a secondary alcohol dehydrogenase of Candida parapsilosis [10c]. Using 2-propanol as an energy source to regenerate NADH, the yield of (R)-ECHB reached 36.6 g l1 (more than 99% ee, 95.2% conversion yield) without addition of NADH to the reaction mixture (Figure 8.14c). On the other hand, a novel carbonyl reductase (KLCR1) that reduced ECAA to (S)-ECHB was puriﬁed from Kluyveromyces lactis [10d]. KLCR1 catalyzed the NADPH-dependent reduction of ECAA enantioselectively, but not the oxidation of (S)-ECHB. 8.3.2.3 Coexpression of Genes for Carbonyl Reductase and Cofactor-Regenerating Enzymes A new bioreduction system for the production of chiral alcohol using an E. coli transformant coexpressing genes for carbonyl reductase and cofactor-regenerating enzymes has been investigated [11]. An NADPH-dependent carbonyl reductase (S1) isolated from C. magnoliae, which catalyzed the reduction of ethyl 4-chloro-3oxobutanoate (COBE) to ethyl (S)-4-chloro-3-hydroxybutanoate (CHBE), with a 100% ee was used to construct the system. The S1 gene comprises 849 bp and encodes a polypeptide of 30 420 Da, and the deduced amino acid sequence showed a high degree of similarity to those of the other members of the short-chain

j 8 Enzymatic Reduction Reaction

204

alcohol dehydrogenase superfamily [11b]. The E. coli cells expressing both the carbonyl reductase gene and the glucose dehydrogenase (GDH) gene from Bacillus megaterium were used as the catalyst for reduction of ethyl 4-chloro-3-oxobutanoate to ethyl (S)-4-chloro-3-hydroxybutanoate [11c]. In an organic-aqueous two-phase system, 450 g l1 of the product was obtained in 89% yield and 100% ee. The system was also applied for the reduction of analogous substrates affording D-alcohols in high enantiomeric excess (Figure 8.15) [11d]. 8.3.2.4 Modification of Biocatalysts: Directed Evolution Directed evolution of enzymes has been used to improve the reducing function of the enzymes. For example, this method was used to eliminate the cofactor requirement of B. stearothermophillus lactate dehydrogenase, which is activated in the presence of fructose 1,6-bisphosphate [12]. The activator is expensive and representative of the sort of cofactor complications that are undesirable in industrial processes. Three rounds of random mutagenesis and screening produced a mutant that is almost fully

Carbonyl reductase (S1) gene

S1 Plasmid

Gene of Candida magnoliae

S1

More effective !!! Glucose dehydrogenase (GDH) gene for cofactor-recycling S1 Plasmid

Overexpression

GDH E. coli Transformant cells for biocatalysis

Gene of Bacillus megaterium GDH O Organic phase (butyl acetate)

OH

X

CO2 Et

X

CO2 Et

(S)

X = Cl Yield 89% 2.70 M (450 gl–1) ee 100%

S1

Aqueous phase E. coli

HO

NADPH

GDH

O

HO O HO OH D-Gluconolactone

X = Cl, Br, I, N3, OH, H ee> 99%

NADP+

HO HO HO

O OH D-Glucose

Figure 8.15 Coexpression of genes for carbonyl reductase and cofactor-regenerating enzymes [11c,d].

OH

8.3 Methodology for Stereochemical Control

j205

activated in the absence of fructose 1,6-bisphosphate. The Km of the enzyme without the cofactor was improved from 5 to 0.07 mM for the pyruvate (Figure 8.16). 8.3.3 Modification of Substrates

The enantioselectivity of a biocatalytic reduction can be controlled by modifying the substrate because the enantioselectivity of the reduction reaction is profoundly affected by the structure of the substrate[13]. For example, in the reduction of 4-chloro3-oxobutanoate by bakers yeast, the length of ester moiety controlled the stereochemical course of the reduction [13a–c]. When the ester moiety was smaller than a butyl group, (S)-alcohols were obtained, and when it was larger than a pentyl group, (R)alcohols were obtained, as shown in Figure 8.17a. After reduction, the ester moiety can be exchanged easily without racemization, so both enantiomers of an equivalent synthetic building block could be obtained using the same reaction system. Another example is the introduction of sulfur functionalities into the keto esters at the a- or a0 positions, which can be eliminated after the reduction. Methylthio1

O

OH OR

bsLDH

O

OR O

70-fold activation !!! Evoluted enzyme

Evolution Evolution Native enzyme (bsLDH)

Km pyr = 5 mM Extraction of gene

Step1 Mutation of gene

Point mutation !

STEP4 Collection of Mutant Gene

(R118C, Q203L, After three rounds N307S)

Evolution cycle

Step2 Expression of mutant enzymes

Figure 8.16 Elimination of the cofactor requirement by directed evolution [12].

Km pyr = 0.07 mM

Step3 Selection of mutant enzyme

j 8 Enzymatic Reduction Reaction

206

R

(a)

n = 5 –12 OH

Cl

CO2(CH2)nH

Cl

Baker’s yeast O CO2(CH2)nH

ee (R)

Baker’s yeast ee (S)

OH Cl

CO2(CH2)nH S

n

n=1–4 OH

(b)

CO2Me OH

SR CO2Me

Baker’s yeast

S

ee > 96%

O

Baker’s yeast ee 87%

CO2Me

O

SR CO2Me O PhO2S

CO2Me Baker’s yeast

OH

ee 98% CO2Me

R

OH PhO2S

CO2Me

Figure 8.17 Modification of the substrate to control the enantioselectivities: (a) effect of the length of the ester moiety1 and (b) effect of introducing sulfur functionalities [13d,e].

[12] and phenylsulfonyl1 [13e] groups were used to improve the enantioselectivities, as shown in Figure 8.17d,e. 8.3.4 Modification of Reaction Conditions 8.3.4.1 Acetone Treatment of the Cell A dried cell mass is often used as a biocatalyst for a reduction since it can be stored for a long time and used whenever needed, without cultivation. One convenient method of drying the cell mass is acetone dehydration. For example, dried cells of G.

8.3 Methodology for Stereochemical Control OH

Untreated whole cell

Ph

O

Dried cell of G. candidum (APG4) Ph

ee 28% (R) Catalyst Untreated whole cell Acetone dried cell (APG4) Acetone dried cell (APG4) Acetone dried cell (APG4) Acetone dried cell (APG4) Acetone dried cell (APG4) Acetone dried cell (APG4)

Coenzyme None None NAD + None NAD + NAD + + NADP

+

+

NAD or NADP 2-Propanol or cyclopentanol Additive

Yield (%)

None None None 2-Propanol 2-Propanol Cyclopentanol Cyclopentanol

52 0 1 8 89 97 86

j207

OH Ph

ee > 99% (S) ee (%) 28 (R) 71 (S) 98 (S) > 99 (S) > 99 (S) > 99 (S)

Figure 8.18 Acetone treatment of Geotrichum candidum for the improvement of enantioselectivity [14].

candidum IFO 4597 can be obtained easily by mixing the cells with cold acetone (20 C) followed by ﬁltering and drying under reduced pressure. Cell drying not only aids the preservation of the cell but also contributes to the stereochemical control as shown in Figure 8.18. The reduction of acetophenone catalyzed by untreated, wet whole cell of G. candidum IFO 4597 resulted in poor enantioselectivity (28% ee(R)). When the form of the catalyst was changed from wet whole-cell to dried powdered-cell (APG4), no reduction was observed, which would indicate the loss of the necessary coenzyme(s) and/or coenzyme regeneration system(s) during the treatment of the cells with acetone. The addition of coenzyme NADþ did not have a signiﬁcant effect on the yield. Addition of 2-propanol resulted in only a small increase in the yield, but a signiﬁcant improvement in the enantioselectivity was observed. Surprisingly, the addition of both NADþ and 2-propanol profoundly enhanced both chemical yield and enantiomeric excess. Furthermore, the addition of NADH, NADPþ, or NADPHinstead of NADþ and the addition of cyclopentanol instead of 2-propanol also gave enantiomerically pure alcohol in high yield. The improvement in the enantioselectivity from 28% (R) to >99% (S) was due to the suppression of every enzyme that reduces the substrate, followed by the stimulation of an S-directing enzyme by the addition of the coenzyme and an excess amount of 2propanol, agents which push the equilibrium toward the reduction of the substrate. It was conﬁrmed, by separating the enzymes in the powder, that many S- and R-directing enzymes do indeed exist in the dried cells. The addition of a coenzyme and cyclopentanol stimulates only an S enzyme because the speciﬁc S enzyme can oxidize cyclopentanol (concomitantly reducing NAD(P)þ), while other S or R enzymes cannot use cyclopentanol as effectively [14c]. This presents a very interesting case, where the experimental conditions of reduction with a cell having both Sand R-directing enzymes was modiﬁed and resulted in excellent S enantioselectivity. 8.3.4.2 Selective Inhibitor In case poor overall enantioselectivity is observed owing to the presence of two competing enzymes with different enantioselectivities, one of the most straightfor-

j 8 Enzymatic Reduction Reaction

208

S R

OH

O CO2R

Additive none

O

OH CO2R

CO2R Baker's yeast

Yield 61% ee 4% R = Et Yield 56% ee 96% R = Me

Additive O Cl

OEt Mg2+

Yield 63% ee 94% R = Et Yield 65% ee 99% R = Et

Figure 8.19 Improvement of the enantioselectivity by using an inhibitor of undesired enzymes [15a–c].

ward methods for improving the enantioselectivity is to use an inhibitor of the unnecessary enzyme(s). Ethyl chloroacetate, methyl vinyl ketone, allyl alcohol, allyl bromide, sulfur compounds, Mg2þ, Ca2þ, and so on, have been reported as inhibitors of enzymes in bakers yeast [15]. For example, the low enantioselectivity in the yeast reduction of b-keto ester was improved by addition of methyl vinyl ketone, Mg2þ, or ethyl chloroacetate as described in Figure 8.19 [15a–c]. By enzymatic studies using puriﬁed enzymes from baker's yeast, the enzymes inhibited were identiﬁed, and the inhibition mechanism was reported to be noncompetitive [15k]. 8.3.4.3 Reaction Temperature Reaction temperature is one of the parameters affecting the enantioselectivity of a reaction [16]. For the oxidation of an alcohol, the values of kcat/Km were determined for the (R)- and (S)-stereodeﬁning enantiomers; E is the ratio between them. From the transition state theory, the free energy difference at the transition state between (R) and (S) enantiomers can be calculated from E (Equation 2), and G is in turn the function of temperature (Equation 3). The racemic temperature (Tr) can be calculated as shown in (Equation 4). Using these equations, Tr for 2-butanol and 2-pentanol of the Thermoanaerobacter ethanolicus alcohol dehydrogenase were determined to be 26 and 77 C, respectively.

E ¼ ðkcat =K m ÞR =ðkcat =K m ÞS

ð1Þ

From transition state theory, RT lnðEÞ ¼ DDGz

ð2Þ

DDGz ¼ DDHz TDDSz

ð3Þ

When DDGz ¼ 0;

ð4Þ

T r ¼ DDHz =DDSz

Since the transition state for alcohol oxidation and ketone reduction must be identical, the product distribution (under kinetic control) for reducing 2-butanone and 2-pentanone is also predictable. Thus, one would expect to isolate (R)-2-butanol if the temperature of the reaction was above 26 C. On the contrary, if the temperature is less than 26 C, (S)-2-butanol should result; in fact, the reduction of

8.4 Medium Engineering

j209

2-butanone and 2-pentanone at 37 C resulted in 28% ee (R)- and 44% ee (S) alcohol, respectively, as expected [16a]. 8.4 Medium Engineering

Biocatalytic reduction has been performed in nonaqueous solvents to improve the efﬁciency of the reaction. This section explains the use of organic solvent, supercritical ﬂuids, and ionic liquid. 8.4.1 Organic Solvent 8.4.1.1 Soluble Organic Solvent Soluble organic solvents have often been used as cosolvents to solubilize miscible organic substrates. Since organic compounds including solvents are possibly incorporated inside of the enzyme, they may affect the stereoselectivity of enzymatic reactions. For example, dimethyl sulfoxide (DMSO) (10%) enhance not only chemical yield but also enantioselectivity of yeast reduction. Thus, the poor yield of 23% with 80% ee was increased to 65% yield with >99% ee (Figure 8.20) [17]. 8.4.1.2 Aqueous-organic Two-phase Reaction Aqueous-organic two-phase reaction has been widely performed [18]. One of the purposes of using two-phase reaction system is to control the substrate concentration in aqueous phase where the biocatalysts exist. Hydrophobic substrate and products dissolve easily in the organic phase, so that the concentration in the aqueous phase decreases. The merits of controlling and decreasing the substrate concentration in the aqueous phase are as follows:

1. Substrate and product inhibition can be prevented. 2. When whole cell containing plural enzymes with opposite selectivities and different Km (Michaelis–Menten constant) values are used, problems of low selectivities occur. If the substrate concentration is decreased, one of the enzymes with low Km value catalyzes the reaction so that the selectivity can be improved. 3. The decomposition of unstable substrate/product by aqueous buffer can be prevented by dissolving the substrate and product in the organic phase. Therefore, organic solvents have been widely used for biocatalytic reductions. An interesting example for stereochemical control by using organic solvents for OH

O Cl

Baker's yeast

Figure 8.20 Stereochemicalcontrolbysolubleorganicsolvent[17].

Cl

Yield

ee

Without DMSO

23%

80%

+ 10% DMSO

65% > 99%

j 8 Enzymatic Reduction Reaction

210

the reduction is as follows. G. candidum IFO 4597 catalyzes the reduction of ketones although the reduction of acetophenone, by the resting cell in water, afforded only (R)-alcohol with 28% ee in a 52% yield [18a]. The cell was immobilized on a water-absorbing polymer to spread it on the large surface of the polymer to use it in hexane (Figure 8.21) [18a]. The reduction in hexane by the immobilized cell

Hexane layer

Hexane layer

OH

O

(S)

Reaction in cell

O

OH

NAD(P)H

NAD(P)+

O

OH C4H9

C4H9 Excess amount

Cell

Aqueous layer inside the polymer

Solubility in aqueous layer 1-phenylethanol > acetophenone

Water absorbing polymer O

Geotrichum candidum R

Substrate

Hexane

OH R Yield (%)

52a Acetophenone 2b Acetophenone 73 Acetophenone 81 2-Acetylfuran 99 o-Chloroacetophenone 88 m-Chloroacetophenone 41 p-Chloroacetophenone 59 o-Methylacetophenone 56 m-Methylacetophenone 40 p-Methylacetophenone 97 1-Phenyl-2-propanone 88 Benzylacetone a Reaction in water without immobilization and without 2-hexanol. b Reaction without 2-hexanol.

Figure 8.21 Reduction of acetophenone in hexane by the immobilizedrestingcellofGeotrichumcandidumand2-hexanol[18a].

ee (%) 28 (R)a — >99 (S) 99 (S) 99 (S) >99 (S) 92 (S) >99 (S) >99 (S) >99 (S) >99 (S) >99 (S)

8.4 Medium Engineering

barely proceeded because of the unfavorable partition of ketone and alcohol between the organic and aqueous layers for the reduction (Figure 4.21). However, when an alcohol such as 2-hexanol was added to the reaction system, the reduction proceeded smoothly to afford (S)-alcohol in excellent enantiomeric excess. 2-Hexanol as well as other 2-alkanols from 2-propanol to 2-octanol worked when they were used in large excess and reached to saturation on the yield at 10–15 mol equiv. The mechanism of enantioselectivity improvement by this method can be explained as follows: (i) enantiomeric excess is poor for reduction in water because both the S enzyme(s) and R enzyme(s) catalyze the reduction; (ii) when hexane is used, both enzymes are inhibited; and (iii) when 2-alkanol was added, only an S enzyme was reactivated and catalyzed the reduction (2-alkanols are selective to the S enzyme). Various aryl methyl ketones were also reduced smoothly by the same system using 12 equivalents of 2-hexanol to produce the corresponding (S)-alcohols in excellent enantiomeric excess (Figure 8.21). The reduction of ortho-substituted acetophenones gives a relatively better yield than that of para-substituted acetophenones, probably because the reverse reaction, oxidation, does not proceed for the ortho-substituted substrates. The better a substrate is for oxidation, the worse it is for reduction of the substituted acetophenone derivatives. In another example, methyl 7-ketolithocholate (Me-7KLCA) was reduced with Eubacterium aerofaciens JCM 7790 in anaerobic interface bioreactor and dihexyl ether was used as the solvent [18b,c]. The microbe system reduced 12 g l1 of Me-7KLCA to methylursodeoxycholate (Me-UDCA) in more than 50% yield. The product is the precursor of ursodeoxycholic acid, which is used as a cholesterol gallstone-dissolving agent (Figure 8.22). The enantioselective reduction of alkyl 3-oxobutanoates by carbonyl reductase (Sl) from C. magnoliae was also performed in organic-aqueous two-phase reaction system (Figure 8.15) [11c,d]. 8.4.2 Use of Hydrophobic Resin, XAD

Instead of using organic solvents, hydrophobic resin, AmberliteTM XAD, has been used to control the substrate concentration [19]. When XAD is added to the reaction mixture, substrate and products are adsorbed to the hydrophobic resin, XAD, since Me-7KLCA

Me-UDCA CO2Me

HO

O

CO2Me

HO

OH

Organic phase

Microbial film

Glucose Mannitol

Nutrients Water

Hydrophilic carrier

Figure 8.22 Reduction of ketone with anaerobic interface bioreactor [18b].

j211

j 8 Enzymatic Reduction Reaction

212

G. candidum

O R Substrate

1

R

2

Hydrophobic polymer XAD-7

Without XAD-7 Yield/% ee/%

O C6H13

36

OH

With XAD-7

R

82

> 99 (S) O

86 (R)

82

58

18 (R)

76

C 3 H7

> 99 (S) Ph

O

O

Yield/% ee/%

O

O 19

R2

Substrate Without XAD-7

Yield/% ee/%

60 (S)

1

97 (S)

OPh

With XAD-7 Yield/%

ee/%

14

33 (R)

25

86 (S)

35

94 (S)

32

> 99 (S)

99

3.6 (S)

86

99 (S)

Figure 8.23 Reduction of ketones with Geotrichum candidum in the presence of hydrophobic polymer XAD [19a].

the substrate and product are usually hydrophobic. Therefore, the effective concentration of the substrate around the enzyme is decreased. Recently, XAD was used as material to control the stereochemical course of microbial reductions [19]. In the presence of XAD, simple aliphatic and aromatic ketones were reduced to the corresponding (S)-alcohols in excellent enantioselectivity while low enantioselectivities were observed in the absence of the polymer (Figure 8.23). In the reduction of benzoyloxypropanone, the hydrophobic polymer XAD-7 was used to prevent product inhibition and increase substrate concentration [19b]. Thus, the reduction proceeded in 70 g l1 substrate concentration and afforded 87% (12.4 g) of (S)-1-benzoyloxy-2-propanol in >99% ee (Figure 8.24). The butanone derivative could be reduced with the same method and afforded (S)-alcohol in 72% yield and >99% ee, but the pentanone derivative could not be reduced. An acyclic enone, 2-ethyl-1-phenylprop-2-en-1-one, was reduced with the yeast Pichia stipitis CCT 2617 [19c]. The reduction proceeded chemo- and enantioselectively to afford (S)-2-ethyl-1-phenylprop-2-en-1-ol (65% yield, >99% ee). XAD-7 was used to decrease and control the concentration of both the substrate and the product (Figure 8.25). O

Baker's yeast OBz

XAD-7, ethanol, O 2

OH OBz

(S) Yield 87% ee > 99%

Figure 8.24 Reduction of ketone with bakers yeast in the presence of hydrophobic polymer XAD [16b].

O

OH Pichia stipitis XAD-7

Figure 8.25 Reduction of ketone with Pichia stipitis in the presence of hydrophobic polymer XAD [19c].

(S) Yield 65% ee > 99%

8.4 Medium Engineering

8.4.3 Supercritical Carbon Dioxide

Reduction using alcohol dehydrogenases is usually conducted in aqueous media. The difﬁculties encountered in such reactions are the extraction of products that dissolve in aqueous media at low concentration, and an organic solvent is usually 0 used. However, by using scCO2 this becomes unnecessary because CO2 transforms into a gas as the pressure decreases. Therefore, reduction of carbonyl group using alcohol dehydrogenases was conducted in scCO2. For example, immobilized cells of G. candidum were used for the reduction in scCO2 [20a,b]. Since the whole resting cells were used, the addition of expensive coenzymes was avoided. Moreover, the solubility of the coenzymes in scCO2 was not required to be considered. At ﬁrst, the reduction of o-ﬂuoroacetophenone in scCO2 at 10 MPa was conducted using 2-propanol as a reductant (hydrogen donor) (Figure 8.26), which afforded (S)-1(o-ﬂuorophenyl)ethanol in 81% yield (determined by gas chromatography) after 12 hours [20a]. The time course of the reaction shows that the yield increased with the reaction time, which proved that the alcohol dehydrogenase catalyzed the reduction in the supercritical condition. The substrate speciﬁcity was investigated, and as listed in Figure 8.26b, the enzymatic reduction in scCO2 proceeded for various ketones. Acetophenone, acetophenone derivatives, benzyl acetone, and cyclohexanone were used as substrates, and it was found that all of them were reduced by the alcohol dehydrogenase in scCO2 with 2-propanol. The effects of ﬂuorine substitution at the ortho, para, and a-positions of acetophenone were obvious. Compared with the unsubstituted analog, substitution at the ortho position increased the yield, whereas substitution at the para position decreased the yield. Regarding enantioselectivity, very high values (>99% ee) were obtained for the reduction with the majority of the substrates tested, while slightly lower enantioselectivities (96, 97% ee) were observed for a few of them. The enantioselectivities obtained in this system are superior or at least equal to those for most other biocatalytic and chemical systems. The immobilized resting cell of G. candidum was also used as a catalyst for the reduction of o-ﬂuoroacetophenone and cyclohexanone in a semicontinuous ﬂow process using scCO2 [20b]. The apparatus is shown in Figure 8.26c. With ﬂow reactors, the addition of a substrate to the column with a catalyst yielded the product and CO2, which is a gas at ambient pressure, whereas, with the batch reactor, separation of the product from the biocatalyst was necessary after depressurization. Therefore, the ﬂow type was superior to the batch type for achieving virtually no solvent reaction. Moreover, the size of the reactors using the ﬂow process to generate an amount of product comparable with the corresponding batch reactors is smaller, which is particularly attractive for a supercritical ﬂuid system. This reaction using a semicontinuous ﬂow process also resulted in a higher space-time yield than that of the corresponding batch process. An isolated enzyme is also used for the reduction in scCO2 [20c]. HLADH was used for reduction of butyraldehyde as shown in Figure 8.27. In this case, addition of

j213

j 8 Enzymatic Reduction Reaction

214

Figure 8.26 Asymmetric reduction of ketones in CO2 by Geotrichum candidum immobilized whole cell [20]. (a) Time course for the reduction of o-fluoroacetophenone; (b) substrate specificity; (c) apparatus for G. candidum–catalyzed reduction with semiflow process using scCO2.

8.4 Medium Engineering

(a)

Horse liver alcohol dehydrogenase (HLADH)

O

OH H

H Ethanol, CO2 (17.9 MPa, 21 oC), 45 h

250 mM (b)

NH 2

O NH2

N O

N O O O P O P O OOH

O N NH(CH2 )6 NH N

N

CF O CF 2 CF F CF3 CF3 n

O Fluorinated polymer attached

OH

OH

Yield (mM)

(c)

OH

OH

50 40

Coenzymes 1: No added coenzyme 2: NAD+ 3: Fluorinated NAD+

30 20 10 0

4: HLADH lyophilized with NAD+ 1

2 3 Coenzymes

4

Figure 8.27 Reduction of aldehyde in scCO2 by an isolated enzyme, horse liver alcohol dehydrogenase (HLADH) [20c] (a) Reaction scheme; (b) fluorinated coenzyme soluble in CO2; and (c) effect of coenzyme on the reaction.

coenzyme is necessary, and the sample with a soluble coenzyme, ﬂuorinated coenzyme (Figure 8.27b), demonstrated the highest activity (Figure 8.27c). The sample with enzyme alone and the sample with added NADþ showed similar rate of butanol production. NADþ added separately from the enzyme did not change native HLADH activity because the enzyme and NADþ remained separate during the reaction. The HLADH lyophilized with NADþ produced very little butanol owing to mass transfer limitations. 8.4.4 Ionic Liquid

Ionic liquid [bmim]PF6 can be used as a solvent in yeast reduction [21]. The reduction of ketones with immobilized bakers yeast (alginate) in a 100 : 10 : 2 [bmim]PF6 ionic liquid: water: MeOH mix affords chiral alcohols (Figure 8.28).

j215

j 8 Enzymatic Reduction Reaction

216

OH (S) Yield 40%, ee 79% O R1

OH

Baker's yeast (immobilized with alginate) R2 N + N

Bu

(S) Yield 22%, ee 95%

OH R1

Me

O

R2

PF6–

OH OH

CO2Et (S) Yield 70%, ee 95% CO2Et

[bmim]PF6 : H2O: MeOH = 100 : 10 : 2 OH CO2Et

(S) Yield 75%, ee 84% (S) Yield 60%, ee 76%

Figure 8.28 Use of ionic liquid in yeast reduction [21].

8.5 Synthetic Applications 8.5.1 Reduction of Aldehyde

Many aldehyde reductases transform both aldehydes and ketones. For example, phenylacetaldehyde reductase (PAR) from a styrene-assimilating Corynebacterium strain, ST-10, reduces hexyl aldehyde and phenylacetaldehyde [22a]. Other aldehyde reductases such as one from Sporobolomyces salmonicolor also reduce aldehydes as well as ketones [22b]. Organometallic aldehydes can be reduced enantioselectively with dehydrogenases. For example, optically active organometallic compounds having planar chiralities were obtained by biocatalytic reduction of racemic aldehydes with yeast [22c,d] or HLADH [22e] as shown in Figure 8.29. 8.5.2 Reduction of Ketone

Simple aliphatic ketones as well as aromatic ketones can be reduced with very high enantioselectivity by using biocatalysts [14]. For example, aliphatic ketones such as 2-pentanone, 2-butanone, 3-hexanone, and so on, were reduced with excellent enantioselectivity to the corresponding (S)-alcohols by using the dried cells of G. candidum (Figure 8.30) [14e]. The dried-cell G. candidum system can distinguish between two alkyl groups with a difference of a single methylene unit. The dried-cell G. candidum system is also applied for the reduction of aromatic ketones. Triﬂuoromethyl ketones were also reduced by the same system and afford (S)-alcohols in excellent enantiomeric excess [14h]. The substrate speciﬁcity of the system is shown (Figure 8.30) [14].

8.5 Synthetic Applications

O

O

Yeast

CH2OH Fe(CO) 3

H

Fe(CO) 3

j217

+ Fe(CO) 3

Yield 53% ee 78% (S)

Yield 32% ee > 99% (R)

CH2OH

CHO

H

CHO

HLADH +

NAD + EtOH Cr(CO) 3

Cr(CO) 3

Cr(CO) 3

Yield 33% ee 91%(S)

Yield 51% ee 81%(R)

Figure 8.29 Reduction of organometallic aldehydes to produce alcohols with planar chiralities [22c,e].

Resting cell of G. candidum, as well as dried cell, has been shown to be an effective catalyst for the asymmetric reduction. Both enantiomers of secondary alcohols were prepared by reduction of the corresponding ketones with a single microbe [23]. Reduction of aromatic ketones with G. candidum IFO 5767 afforded the corresponding (S)-alcohols in an excellent enantioselectivity when amberlite XAD-7, a hydrophobic polymer, was added to the reaction system, and the reduction with the same microbe afforded (R)-alcohols, also in an excellent enantioselectivity, when the reaction was conducted under aerobic conditions (Figure 8.31). O

OH R

H3 C

CF3

X3C R Dried cell of G. candidum (APG4) 2-propanol or cyclopentanol

F3C

R

Product OH

OH

OH

OH

OH

OH Cl

94% ee

>99% ee

OH

98% ee

OH Ph

> 99% ee

OH

OH F3C > 99% ee

> 99% ee

OH F3C

Ph

98% ee

OH CO2Et

> 99% ee

OH

OH F3 C Cl

Ph

> 99% ee

> 99% ee

F3C

Figure 8.30 Reduction of ketones by the dried cell of G. candidum, NAD(P)þ, and secondary alcohol [14].

OH MeO

Ar

OH

> 99% ee

>99% ee

99% ee

OH CO2Et

Ph > 99% ee

99% ee

S

>99% ee

F3 C 98% ee

Ph

j 8 Enzymatic Reduction Reaction

218

OH

O

XAD-7

R

OH

Aerobic conditions R

R G. candidum IFO5767

(S)

(R)

92–> 99% ee

85–> 99% ee

96–> 99% yield

61–> 99% yield

O

O

O

O

O

O

O

N N

Br

Figure 8.31 Synthesis of both enantiomers of secondary alcohols with one kind of microbe [23].

Bakers yeast has been widely used for the reduction of ketones. The substrate speciﬁcity and enantioselectivity of the carbonyl reductase from bakers yeast, which is known to catalyze the reduction of b-keto ester to L-hydroxyester (L2-enzyme) [15], was investigated, and the enzyme was found to reduce chloro-, acetoxy ketones with high enantioselectivity (Figure 8.32) [24a]. The reduction of hydroxy or acetoxy ketones by bakers yeast shows an interesting stereoselectivity. For the reduction of acetylbenzofuran derivatives with bakers yeast, the methyl ketones afforded (S)-alcohol in 20–68% ee. The hydroxyl derivatives afforded (S)-alcohol in 87–93% ee, and the acetoxy derivatives gave (R)-alcohols in 84–91% ee (Figure 8.33) [24b]. Reductase from baker's yeast O R

OH Cl

(R), ee > 99%

1

(R), ee 88%

OH

OH

OH

Ph Cl

R2

R

glucose-6P glucose-6PDH

OH Cl

OH

NADP+ 2

R1

OAc

Ph (R), ee >99%

(R), ee 98%

Ph

OAc

(R), ee 96%

Figure 8.32 Reduction of ketones with reductase from bakers yeast [24].

OH X

O

HO

O O

Baker's yeast

R

R X = OH or OAc R = H, 5-Br, 5-NO2, 7-MeO

(S) X = OH: ee 87–93% OH AcO

OH O

HO R

Figure 8.33 Reduction of hydroxy and acetoxy ketones [24b].

O

(R) X = OAc: ee 84–91%

R

8.5 Synthetic Applications

OH

O

OH

R1

R1 R2

R1 R2

(S) NADH CPCR

R2

(R)

NAD+ R1 = CH3, R2 = Ar : ee > 99% (S)

CO2 FDH

HCO2H

R1 = CH3, R2 = SiMe3 : ee > 99% (S) R1 = CH3, R2 = H : ee > 60% (R) R1 = C3H7, R2 = H : ee > 99% (S)

Figure 8.34 Reduction of acetylenic ketones [25].

For reduction of acetylenic ketones, two oxidoreductases were used [25]. Lactobacillus brevis alcohol dehydrogenase (LBADH) gave the (R)-alcohols and Candida parapsilosis carbonyl reductase (CPCR) afforded the (S)-isomer, both in good yield and excellent enantioselectivity. By changing the steric demand of the substituents, the enantiomeric excess values can be adjusted and even the conﬁgurations of the products can be altered (Figure 8.34). Process with PAR produced by styrene-assimilating Corynebacterium strain ST-10 is one of the best asymmetric reductions ever reported to synthesize chiral alcohols [26]. This enzyme with a broad substrate range reduced various prochiral aromatic ketones and b-keto esters to yield optically active secondary alcohols with an enantiomeric purity of more than 98% ee. The E. coli recombinant cells, which expressed the PAR gene, could efﬁciently produce important pharmaceutical intermediates: (R)-2-chloro-1-(3-chlorophenyl)ethanol (28 mg ml1) from m-chlorophenacyl chloride, ethyl (R)-4-chloro-3-hydroxy butanoate) (28 mg ml1) from ethyl 4-chloro-3oxobutanoate, and (S)-N-tert-butoxycarbonyl(Boc)-3-pyrrolidinol from N-Boc-3pyrrolidinone (51 mg ml1), with more than 86% yields. The high yields were due to the fact that PAR could concomitantly reproduce NADH in the presence of 3–7% (v/ v) 2-propanol in the reaction mixture (Figure 8.35). Phenylacetaldehyde reductase from Corynebacterium strain ST-10 (Escherichia coli recombinant cells)

O R1

R2

OH R1

2-Propanol

OH

R2 OH

OH

Cl

Cl

OMe

(R), ee > 99%

OH

OH

OH CO2Et

(S), ee > 99%

Cl

OH CO2Et

(S), ee > 99%

Cl (S), ee > 99%

OMe (S), ee > 99%

(S), ee > 99%

Cl

OMe

Br

HO CO2Et

(S), ee 98.4%

Figure 8.35 Reduction of ketones with phenylacetaldehyde reductase from Corynebacterium strain ST-10 [26].

N

CO2But

(S), ee > 99%

j219

j 8 Enzymatic Reduction Reaction

220

(a)

Rhodococcus ruber

O

OH

2-Propanol

R

(S) > 99% ee

R

[27a]

R = alkyl, ph O OH

Rhodococcus ruber

(b)

(S) > 99% ee

2-Propanol

OH

OH

Rapsbery ketone

[27b]

Figure 8.36 Reduction of ketones by Rhodococcus ruber [27].

The biocatalytic process using Rhodococcus ruber DSM 44541 is fully developed. Reduction of ketones with the lyophilized cells using 2-propanol as hydrogen donor afforded (S)-alcohol in high enantiomeric excess (Figure 8.36a) [27a]. Raspberry ketone was also reduced with the biocatalyst and gave (S)-rhododendrol in 87% yield with >99% ee (Figure 8.36b) [27b]. Vegetables could be used as biocatalysts for the reduction of ketones as shown in Figure 8.37 [28]. The reaction of racemic-2-methylcyclohexanone with fresh carrot root [28a] and various vegetables [28b] gave a mixture of (1S,2R)- and (1S,2S)-2-methylcyclohexanol in high enantiomeric excess (Figure 8.37a). The reaction of 2-hydroxycyclohexanone afforded a 1: 2 mixture of (1S,2R)- and (1S,2S)-1,2-cyclohexanediol with >95% ee [28]. For the reduction of 4-phenyl-2-oxo-but-3-enoates, the corresponding (R)-alcohols were obtained in 92–99% ee (Figure 8.37b) [28c]. Prochiral ketones such as aromatic ketones [28d–f] and b-keto esters [28d] were reduced to the corresponding optically active alcohols by carrot and other vegetables (Figure 8.37c) [28d]. O

OH

(a)

(b)

R

Carrot root

O CO2R

X

OH R

R

R = CH3, OH ee > 95%

OH

Plant cell culture of carrot

CO2R

X (R)

(c)

OH

X =H, o-Cl, p-Cl, p-Me R = Me, Et [28c] ee = 92–99%

OH CO2Et

O R1

[28a,b]

R2

OH

Vegetable R1

[28d]

[28d]

R2

OH OH N3

CO2Et

[28d] Figure 8.37 Reduction of ketones by vegetables [28].

X [28d-f]

8.5 Synthetic Applications

8.5.3 Dynamic Kinetic Resolution and Deracemization 8.5.3.1 Dynamic Kinetic Resolution Dynamic kinetic resolution of racemic ketones proceeds through asymmetric reduction when the substrate does racemize and the product does not under the applied experimental conditions. Dynamic kinetic resolution of a-alkyl b-keto ester has been performed through enzymatic reduction. One isomer, out of the four possible products for the unselective reduction (Figure 8.38), can be selectively synthesized using biocatalyst, and by changing the biocatalyst or conditions, all of the isomers can be selectively synthesized [29]. Dynamic kinetic resolution of a-alkyl-b-keto ester was conducted successfully using biocatalysts. For example, baker's yeast gave selectively syn(2R, 3S)-product [29a] and the selectivity was enhanced by using selective inhibitor [29b] or heat treatment of the yeast [29c]. Organic solvent was used for stereochemical control of G. candidum [29d]. Plant cell cultures were used for reduction of 2-methyl-3-oxobutanoate and afforded antialcohol with Marchantia [29e,f ] and syn-isomer with Glycine max [29f ]. Extensive screening methodology was used to ﬁnd the suitable microorganism. As a result, K. pneumoniae IFO 3319 out of 450 bacterial strains was found to give the corresponding (2R, 3S)-hydroxy ester with 99% de and >99% ee in kilogram scale quantitatively [29g]. A series of 2-(4-chlorophenoxy)-3-oxoalkanoate were reduced by bakers yeast, and Kluyveromyces marxianus. Yeast reduction of ethyl 2-(4-chlorophenoxy)-3-oxo-3phenylpropanoate (R ¼ Ph) afforded enantiomerically pure ethyl (2R, 3S)-2(4-chlorophenoxy)-3-hydroxy-3-phenylpropanoate out of the four possible stereoisomers in >99% de [29h]. Although bakers yeast reduction of butanoate (R ¼ CH3) was not selective (92% de), the use of K. marxianus afforded (2R, 3S)-isomer selectively [29i]. The products are intermediates for potential peroxisome proliferator-activated receptor isoform a-agonists (Figure 8.39a). b-Diketones can also be reduced diastereoselectively. Thus, a recombinant alcohol dehydrogenase from L. brevis (recLBADH) overexpressed in E. coli was used for

O

O

OH

O

OEt Fast in aqueous buffer O

O

O

OEt Anti (2R, 3R) OH

OEt

OH

O

OEt Syn (2R, 3S) OH

OEt Syn (2S, 3R)

O OEt

Anti (2S, 3S)

Figure 8.38 Possible products for the reduction of a-methyl b-keto ester.

j221

j 8 Enzymatic Reduction Reaction

222

(a) OH

O CO2Et

Kluyvromyces marxianus

O

OH CO2Et

R O

O

R = Ph

R = CH3 (2R,3S) de > 99%

CO2Et

Ph

Baker's yeast

[29h,i].

Cl

(2R,3S) de > 99%

Cl

(b)

Cl

[29j] O

O CO2But

OH

recLBADH NADP+

O

Yiled 66% CO2But ee 99.2% de 94% (syn)

2-Propanol

OMe

OMe

(c)

S O (S)-ketone

N H

[29k]

S Baker's yeast

(2S, 3S)-Diltiazem

OH N H O Yield 80%, ee > 99% (2S, 3S)

O

OMe S O N (R)-ketone H

O

[29l] Ar

d) O Ar

OH

Curvularia lunata CN

CN

Ar R (R, R)

R

C2H5 Ph C3H7 Ph C4H9 Ph Me2CHCH2 Ph C2H5 m-CH3Ph C2H5 p-CH3Ph C2H5 2-Thienyl C2H5 2-Furyl

Yield de ee (%) (%) (%) 69 38 13 14 42 64 63 59

96 98 86 94 78 92 94 72

98 98 86 97 70 83 93 87

Figure 8.39 Dynamic kinetic resolution.

the reduction of tert-butyl 4-methyl-3,5-dioxohexanoate. The syn-hydroxyester was obtained in 66% yield with 94% de and 99.2% ee (Figure 8.39b) [29j]. Another example of dynamic kinetic resolution is the reduction of a sulfursubstituted ketone. Thus, yeast reduction of (R/S)-2-(4-methoxyphenyl)-1, 5-benzothiazepin-3,4(2H, 5H)-dione gave only (2S, 3S)-alcohol as a product out of four possible isomers as shown in Figure 8.39c [29k]. Only (S)-ketone was recognized by the enzyme as a substrate and reduction of the ketone proceeded

8.5 Synthetic Applications

j223

enantioselectively. The resulting product was used for the synthesis of (2S, 3S)Diltiazem, a coronary vasodilator. White-rot fungus has been used as a biocatalyst for reduction and alkylation. The reaction of aromatic b-keto nitriles with the white-rot fungus Curvularia lunata CECT 2130 in the presence of alcohols afforded alkylation–reduction reaction [29l]. Alcohols such as ethanol, propanol, butanol, and isobutanol could be used (Figure 8.39d). The dynamic resolution of an aldehyde is shown in Figure 8.40. The racemization of starting aldehyde and enantioselective reduction of carbonyl group by bakers yeast resulted in the formation of chiral carbon centers. The enantiomeric excess value of the product was improved from 19 to 90% by changing the ester moiety from the isopropyl group to the neopentyl group [30a]. Other biocatalysts were also used to perform the dynamic kinetic resolution through reduction. For example, Thermoanaerobium brockii reduced the aldehyde with a moderate enantioselectivity [30b,c], and Candida humicola was found, as a result of screening from 107 microorganisms, to give the (R)-alcohol with 98.2% ee when ester group was methyl [30d]. 8.5.3.2 Deracemization through Oxidation and Reduction Deracemization reaction, which converts the racemic compounds into chiral form in one step in one pot without changing their chemical structures can be performed using microorganism containing several different stereochemical enzymes (Figure 8.41) [31]. For example, for deracemization of 1,2-pentandiol (Figure 8.41a), the (R)-speciﬁc NADH-enzyme in C. parapsilosis was reversible and (S)-speciﬁc NADPH-enzyme in the same microorganisms was irreversible [31a]. Therefore, whole-cell reaction of racemic 1, 2-pentandiol gave (S)-diol in 93% yield and 100% ee. For the deracemization of phenylethanol derivatives using G. candidum under aerobic conditions (Figure 8.41b), the (S)-speciﬁc enzyme was reversible and (R) enzyme was irreversible, so (R)-alcohol accumulated when the cell and racemic alcohols were mixed [31b,c]. Para-substituted phenylethanol derivatives gave better results than meta-substituted derivatives. Sphingomonas was used for

(e)

O CO2R

H

(R)-aldehyde

CO2R

(R)-alcohol

Biocatalysts Baker's yeast Baker's yeast Baker's yeast Baker's yeast

R

Yield (%)

70–80 –CH2CH3 49 –CH(CH 3)2 –CH 2CH(CH3)2 84 78 –CH 2C(CH3)3

Thermoanaerobium –CH 2CH3 brockii

O H

HO

ee (%) Reference

60–65 19 64 90

[30b] [30a] [30a] [30a]

50–80

72

[30c]

— —

98.2 73.6

[30d] [30d]

CO2R Candida humicola Candida humicola

(S)-aldehyde

Figure 8.40 Dynamic kinetic resolution of aldehydes.

–CH 3 –CH 2CH3

j 8 Enzymatic Reduction Reaction

224 (a)

NADH dependent R-specific alcohol dehydrogenase

[31a]

NADPH dependent S-specific reductase

OH

OH

OH (R)

Yield 93% ee 100% (S)

OH

O

OH (S)

Candida parapsilosis

[31b,c] (b) OH

X

X

(S)

X

OH

O

H F Cl Br Me MeO

X (R)

Geotrichum candidum Aerobic conditions

ee (%) Configuration

Yield of alcohol (%) 96 96 93 99 99 65

99 100 98 100 96 100

(c) OH

Sphingomonas paucimobilis

S N

[31c]

OH Yield 85% ee 97% (R)

S N OH

OH

(d)

R R R R R R

[31d] Yield 93% ee 100%

* N OH

N Catharanthus roseus plant cell culture OH Yield 100% ee 87% (R)

N N

Figure 8.41 Deracemization via oxidation–reduction.

deracemization of a thiazole derivative (Figure 8.41c) [31d]. Plant cell culture could be used for deracemization of aromatic alcohols (Figure 8.41d) [31e]. 8.6 Conclusions

Asymmetric reduction of carbonyl groups by biocatalysts has been a useful method for preparation of valuable compounds as shown in this review. Increasingly, bioengineering technology has been applied to alcohol dehydrogenases and reductases as has been applied to hydrolytic enzymes to improve enzyme stability, reactivity and enantioselectivity and to extend substrate speciﬁcity. Novel enzymes for reductions created by this technique will be available in large quantities and varieties within a next few years. In the near future, a lot of useful biocatalysts for reduction will be on the market, and expanding number of chemists can use enzymes for reduction more freely than at present owing to the simpliﬁcation of experimental procedures. Furthermore, the biocatalysts will be even more important with the shift of the raw materials from oil to biomass. Since biomass is a mixture of various multifunctional compounds, chemo-, regio-, and enantioselective catalysts will be

References

indispensable, and biocatalysts will play an important role. Moreover, biocatalysts will perform better using natural substrates from biomass than using manmade substrates.

References 1 (a) Nakamura, K., Yamanaka, R., Matsuda, T. and Harada, T. (2003) Tetrahedron: Asymmetry, 14, 2659–2681. (b) van der Donk, W.A. and Zhao, H. (2003) Current Opinion in Biotechnology, 14, 421–426. (c) Nakamura, K. and Matsuda, T. (2002) in Enzyme Catalysis in Organic Synthesis A comprehensive Handbook (eds K. Drauz and H. Waldmann), Wiley-VCH Verlag GmbH, Weinheim, pp. 991–1047. (d) Santaniello, E., Ferraboschi, P. and Manzocchi, A. ,(2000) in Enzymes in Action (eds B. Zwanenburg, P. Mikolajczyk and M. Kielbasinski), Kluwer Academic Publishers, Dordrecht, pp. 95–115. 2 (a) Matsuda, T., Harada, T., Nakajima, N. and Nakamura, K. (2000) Tetrahedron Letters, 41, 4135–4138. (b) Matsuda, T., Harada, T., Nakajima, N., Itoh, T. and Nakamura, K. (2000) Journal of Organic Chemistry, 65, 157–163. (c) Nakamura, K. and Matsuda, T. (1998) Journal of Organic Chemistry, 63, 8957–8964. 3 Iyer, R.B. and Bachas, L.G. (2004) Journal of Molecular Catalysis B: Enzymatic, 28, 1–5. 4 (a) Gr€ oger, H., Hummel, W., Rollmann, C., Chamouleau, F., H€ usken, H., Werner, H., Wunderlich, C., Abokitse, K., Drauz, K. and Buchholz, S. (2004) Tetrahedron, 60, 633–640. (b) Shorrock, V.J., Chartrain, M. and Woodley, J.M. (2004) Tetrahedron, 60, 781–788. 5 (a) Mertens, R., Greiner, L., van den Ban, E.C.D., Haaker, H.B.C.M. and Liese, A. (2003) Journal of Molecular Catalysis B: Enzymatic, 24–25, 39–52. (b) Williams, P., Mickleﬁeld, J., Gardiner, J.M. and Stephens, G. (2004) Tetrahedron, 60, 753–758. (c) Wagenknecht, P.S., Penney, J.M. and Hembre, R.T. (2003) Organometallics, 22, 1180–1182.

6 (a) Nakamura, K., Yamanaka, R., Tohi, K. and Hamada, H. (2000) Tetrahedron Letters, 41, 6799–6802. (b) Nakamura, K. and Yamanaka, R. (2002) Chemical Communications, 1782–1783. (c) Nakamura, K. and Yamanaka, R. (2002) Tetrahedron: Asymmetry, 13, 2529–2533. (d) Utsukihara, T., Chai, W., Kato, N., Nakamura, K. and Horiuchi, C.A. (2004) Journal of Molecular Catalysis B: Enzymatic, 31, 19–24. (e) Utsukihara, T., Misumi, O., Kato, N., Kuroiwa, T. and Horiuchi, C.A. (2006) Tetrahedron: Asymmetry, 17, 1179–1185. (f) Itoh, K., Sakamaki, H., Nakamura, K. and Horiuchi, C.A. (2005) Tetrahedron: Asymmetry, 16, 1403–1408. (g) Parkinson, B.A. and Weaver, P.F. (1984) Nature, 309, 148–149. (h) Miyatani, R. and Amao, Y. (2004) Journal of Molecular Catalysis B: Enzymatic, 27, 121–125. 7 (a) Kim, S., Yun, S.-E. and Kang, C. (1999) Journal of Electroanalytical Chemistry, 465, 153–159. (b) Hollmann, F., Schmid, A. and Steckhan, E. (2001) Angewandte Chemie International Edition, 40, 169–171. (c) Delecouls-Servat, K., Bergel, A. and Basseguy, R. (2001) Journal of Applied Electrochemistry, 31, 1095–1101. (d) Delecouls-Servat, K., Bergel, A. and Basseguy, R. (2004) Bioprocess Biosystems Engineering, 26, 205–215. (e) DelecoulsServat, K., Basseguy, R. and Bergel, A. (2002) Chemical Engineering Science, 57, 4633–4642. (f) Suye, S., Aramoto, Y., Nakamura, M., Tabata, I. and Sakakibara, M. (2002) Enzyme and Microbial Technology, 30, 139–144. 8 (a) Demirjian, D.C., Shah, P.C. and Moris-Varas, F. (2000) Biocatalysis from Discovery to Application (ed. W.-D., Fessner), Springer-Verlag, Berlin,

j225

j 8 Enzymatic Reduction Reaction

226

pp. 1–29. (b) Carballeira, J.D., Álvarez, E., Campillo, M., Pardo, L. and Sinisterra, J.V. (2004) Tetrahedron: Asymmetry, 15, 951–962. (c) Yasohara, Y., Kizaki, N., Hasegawa, J., Takahashi, S., Wada, M., Kataoka, M. and Shimizu, S. (1999) Applied Microbiology and Biotechnology, 51, 847–851. (d) Miya, H., Kawada, M. and Sugiyama, Y. (1996) Bioscience Biotechnology and Biochemistry, 60, 95–98. (e) Shaﬁee, A., Motamedi, H. and King, A. (1998) Applied Microbiology and Biotechnology, 49, 709–717. (f) Kambourakis, S. and Rozzell, J.D. (2004) Tetrahedron, 60, 663–669. (g) Homann, M.J., Vail, R.B., Previte, E., Tamarez, M., Morgan, B., Dodds, D.R. and Zaks, A. (2004) Tetrahedron, 60, 789–797. 9 (a) Kaluzna, I., Andrew, A.A., Bonilla, M., Martzen, M.R. and Stewart, J.D. (2002) Journal of Molecular Catalysis B: Enzymatic, 17, 101–105. (b) Kaluzna, I.A., Matsuda, T., Sewell, A.K. and Stewart, J.D. (2004) Journal of the American Chemical Society, 126, 12827–12832. 10 (a) Yoshizumi, A., Wada, M., Takagi, H., Shimizu, S. and Nakamori, S. (2001) Bioscience Biotechnology and Biochemistry, 65, 830–836. (b) Patel, R.N., Goswami, A., Chu, L., Donovan, M.J., Nanduri, V., Goldberg, S., Johnston, R., Siva, P.J., Nielsen, B., Fan, J., He, W.X., Shi, Z., Wang, K.Y., Eiring, R., Cazzulino, D., Singh, A., Mueller, R. (2004) Tetrahedron: Asymmetry, 15, 1247–1258. (c) Yamamoto, H., Matsuyama, A. and Kobayashi, Y. (2002) Bioscience Biotechnology and Biochemistry, 66, 481–483. (d) Yamamoto, H., Kimoto, N., Matsuyama, A. and Kobayashi, Y. (2002) Bioscience Biotechnology and Biochemistry, 66, 1775–1778. 11 (a) Kataoka, M., Kita, K., Wada, M., Yasohara, Y., Hasegawa, J. and Shimizu, S. (2003) Applied Microbiology and Biotechnology, 62, 437–445. (b) Yasohara, Y., Kizaki, N., Hasegawa, J., Wada, M., Kataoka, M. and Shimizu, S. (2000) Bioscience Biotechnology and Biochemistry,

64, 1430–1436. (c) Kizaki, N., Yasohara, Y., Hasegawa, J., Wada, M., Kataoka, M. and Shimizu, S. (2001) Applied Microbiology and Biotechnology, 55, 590–595. (d) Yasohara, Y., Kizaki, N., Hasegawa, J., Wada, M., Kataoka, M. and Shimizu, S. (2001) Tetrahedron: Asymmetry, 12, 1713–1718. 12 Allen, S.J. and Holbrook, J.J. (2000) Protein Engineering, 13, 5–7. 13 (a) Zhou, B.-N., Gopalan, A.S., VanMiddlesworth, F., Shieh, W.-R. and Sih, C.J. (1983) Journal of the American Chemical Society, 105, 5925–5926. (b) Chen, C.-S., Zhou, B.-N., Girdaukas, G., Shieh, W.-R., VanMiddlesworth, F., Gopalan, A.S. and Sih, C.J. (1984) Bioorganic Chemistry, 12, 98–117. (c) Shieh, W.-R., Gopalan, A.S. and Sih, C.J. (1985) Journal of the American Chemical Society, 107, 2993–2994. (d) Fujisawa, T., Itoh, T. and Sato, T. (1984) Tetrahedron Letters, 25, 5083–5086. (e) Nakamura, K., Ushio, K., Oka, S., Ohno, A. and Yasui, S. (1984) Tetrahedron Letters, 25, 3979–3982. 14 (a) Nakamura, K., Kitano, K., Matsuda, T. and Ohno, A. (1996) Tetrahedron Letters, 37, 1629–1632. (b) Nakamura, K. and Matsuda, T. (1998) Journal of Organic Chemistry, 63, 8957–8964. (c) Matsuda, T., Harada, T., Nakajima, N. and Nakamura, K. (2000) Tetrahedron Letters, 41, 4135–4138. (d) Hamada, H., Miura, T., Kumobayashi, H., Matsuda, T., Harada, T. and Nakamura, K. (2001) Biotechnology Letters, 23, 1603–1606. (e) Matsuda, T., Nakajima, Y., Harada, T. and Nakamura, K. (2002) Tetrahedron: Asymmetry, 13, 971–974. (f) Nakamura, K., Matsuda, T., Itoh, T. and Ohno, A. (1996) Tetrahedron Letters, 37, 5727–5730. (g) Nakamura, K., Matsuda, T., Shimizu, M. and Fujisawa, T. (1998) Tetrahedron, 54, 8393–8402. (h) Matsuda, T., Harada, T., Nakajima, N., Itoh, T. and Nakamura, K. (2000) Journal of Organic Chemistry, 65, 157–163. (i) Nakamura, K., Matsuda, T. and Harada, T. (2002) Chirality, 14, 703–708.

References 15 (a) Nakamura, K., Kawai, Y. and Ohno, A. (1990) Tetrahedron Letters, 31, 267–270. (b) Nakamura, K., Inoue, K., Ushio, K., Oka, S. and Ohno, A. (1987) Chemistry Letters, 16, 679–682. (c) Nakamura, K., Kawai, Y., Oka, S. and Ohno, A. (1989) Tetrahedron Letters, 30, 2245–2246. (d) Dahl, A.C., Fjeldberg, M. and Madsen, J. (1999) Tetrahedron: Asymmetry, 10, 551–559. (e) Nakamura, K., Kawai, Y., Oka, S. and Ohno, A. (1989) Bulletin of the Chemical Society of Japan, 62, 875–879. (f) Forni, A., Moretti, I., Prati, F. and Torre, G. (1994) Tetrahedron, 50, 11995–12000. (g) Kim, J.-H. and Oh, W.-T. (1992) Bulletin of the Korean Chemical Society, 13, 2–3. (h) Hayakawa, R., Nozawa, K., Shimizu, M. and Fujisawa, T. (1998) Tetrahedron Letters, 39, 67–70. (i) Ushio, K., Hada, J., Tanaka, Y. and Ebara, K. (1993) Enzyme and Microbial Technology, 15, 222–228. (j) Hayakawa, R., Shimizu, M. and Fujisawa, T. (1997) Tetrahedron: Asymmetry, 8, 3201–3204. (k) Nakamura, K., Kawai, Y., Nakajima, N. and Ohno, A. (1991) Journal of Organic Chemistry, 56, 4778–4783. 16 (a) Pham, V.T., Phillips, R.S. and Ljungdahl, L.G. (1989) Journal of the American Chemical Society, 111, 1935–1936. (b) Zheng, C., Pham, V.T. and Phillips, R.S. (1994) Catalysis Today, 22, 607–620. (c) Tripp, A.E., Burdette, D.S., Zeikus, J.G. and Phillips, R.S. (1998) Journal of the American Chemical Society, 120, 5137–5141. (d) Heiss, C., Laivenieks, M., Zeikus, J.G. and Phillips, R.S. (2001) Journal of the American Chemical Society, 123, 345–346. 17 Li, F., Cui, J., Qian, X., Ren, W. and Wang, X. (2006) Chemical Communications, 865–867. 18 (a) Nakamura, K., Inoue, Y., Matsuda, T. and Misawa, I. (1999) Journal of the Chemical Society, Perkin Transactions 1, 2397–2402. (b) Oda, S., Sugai, T. and Ohta, H. (2001) Journal of Bioscience and Bioengineering, 91, 178–183. (c) Oda, S., Sugai, T. and Ohta, H. (2001) Journal of Bioscience and Bioengineering, 91, 202–207.

19 (a) Nakamura, K., Fujii, M. and Ida, Y. (2000) Journal of the Chemical Society, Perkin Transactions 1, 3205–3211. (b) Kometani, T., Toide, H., Daikaiji, Y. and Goto, M. (2001) Journal of Bioscience and Bioengineering, 91, 525–527. (c) Andrade Concei~ao, G.J., Moran, P.J.S. and Rodrigues, J.A.R. (2003) Tetrahedron: Asymmetry, 14, 43–45. 20 (a) Matsuda, T., Harada, T. and Nakamura, K. (2000) Chemical Communications, 1367–1368. (b) Matsuda, T., Watanabe, K., Kamitanaka, T., Harada, T. and Nakamura, K. (2003) Chemical Communications, 1198–1199. (c) Panza, J.L., Russell, A.J. and Beckman, E.J. (2002) Tetrahedron, 58, 4091–4104. 21 Howarth, J., James, P. and Dai, J.F. (2001) Tetrahedron Letters, 42, 7517–7519. 22 (a) Itoh, N., Morihama, R., Wang, J., Okada, K. and Mizuguchi, N. (1997) Applied and Environment Microbiology, 63, 3783–3788. (b) Kita, K., Fukura, T., Nakase, K., Okamoto, K., Yanase, H., Kataoka, M. and Shimizu, S. (1999) Applied and Environment Microbiology, 65, 5207–5211. (c) Howell, J.A.S., Palin, M.G., Jaouen, G., Top, S., Hafa, H.E. and Cense, J.M. (1993) Tetrahedron: Asymmetry, 4, 1241–1252. (d) Howell, J.A.S., Palin, M.G., Hafa, H.E., Top, S. and Jaouen, G. (1992) Tetrahedron: Asymmetry, 3, 1355–1356. (e) Baldoli, C., Buttero, P.D., Maiorana, S., Ottolina, G. and Riva, S. (1998) Tetrahedron: Asymmetry, 9, 1497–1504. 23 Nakamura, K., Takenaka, K., Fuji, M. and Ida, Y. (2002) Tetrahedron Letters, 43, 3629–3631. 24 (a) Ema, T., Moriya, H., Kofukuda, T., Ishida, T., Maehara, K., Utaka, M. and Sakai, T. (2001) Journal of Organic Chemistry, 66, 8682–8684. (b) Paizsa, C., Tosa, M., Majdik, C., Moldovan, P., Novak, L., Kolonits, P., Marcovici, A., Irimie, F.-D. and Poppe, L. (2003) Tetrahedron: Asymmetry, 14, 1495–1501. 25 Schubert, T., Hummel, W., Kula, M.R. and M€ uller, M. (2001) European Journal of Organic Chemistry, 4181–4187.

j227

j 8 Enzymatic Reduction Reaction

228

26 Itoh, N., Matsuda, M., Mabuchi, M., Dairi, T. and Wang, J. (2002) European Journal of Biochemistry, 269, 2394–2402. 27 (a) Stampfer, W., Kosjek, B., Faber, K. and Kroutil, W. (2003) Journal of Organic Chemistry, 68, 402–406. (b) Kosjek, B., Stampfer, W., van Deursen, R., Faber, K. and Kroutil, W. (2003) Tetrahedron, 59, 9517–9521. 28 (a) Baldassarre, F., Bertoni, G., Chiappe, C. and Marion, F. (2000) Journal of Molecular Catalysis B: Enzymatic, 11, 55–58. (b) Utsukihara, T., Watanabe, S., Tomiyama, A., Chai, W. and Horiuchi, C.A. (2006) Journal of Molecular Catalysis B: Enzymatic, 41, 103–109. (c) Baskar, B., Ganesh, S., Lokeswari, T.S. and Chadha, A. (2004) Journal of Molecular Catalysis B: Enzymatic, 27, 13–17. (d) Yadav, J.S., Nanda, S., Reddy, P.T. and Rao, A.B. (2002) Journal of Organic Chemistry, 67, 3900–3903. (e) Mczka, W.K. and Mironowicz, A. (2004) Tetrahedron: Asymmetry, 15, 1965–1967. (f) Andrade, L.H., Utsunomiya, R.S., Omori, A.T., Porto, A.L.M. and Comasseto, J.V. (2006) Journal of Molecular Catalysis B: Enzymatic, 38, 84–90. 29 (a) Nakamura, K., Kawai, Y., Miyai, T. and Ohno, A. (1990) Tetrahedron Letters, 31, 3631–3632. (b) Nakamura, K., Kawai, Y., Nakajima, N., Miyai, T., Honda, T. and Ohno, T. (1991) Bulletin of the Chemical Society of Japan, 64, 1467–1470. (c) Nakamura, K., Kawai, Y. and Ohno, A. (1991) Tetrahedron Letters, 32, 2927–2928. (d) Nakamura, K., Takano, S. and Ohno, A. (1993) Tetrahedron Letters, 34, 6087–6090. (e) Speicher, A., Roeser, H. and Heisel, R. (2003) Journal of Molecular Catalysis B: Enzymatic, 22, 71–77. (f) Nakamura, K., Miyoshi, H., Sugiyama, T. and Hamada, H. (1995) Phytochemistry, 40, 1419–1420. (g) Miya, H., Kawada, M. and Sugiyama, Y.

(1996) Bioscience Biotechnology and Biochemistry, 60, 95–98. (h) Perrone, M.G., Santandrea, E., Scilimati, T., Tortorella, V., Capitelli, F. and Bertolasi, V. (2004) Tetrahedron: Asymmetry, 15, 3501–3510. (i) Perrone, M.G., Santandrea, E., Scilimati, A., Syldatk, C., Tortorella, V., Capitelli, F. and Bertolasi, V. (2004) Tetrahedron: Asymmetry, 15, 3511–3517. (j) Ji, A., Wolberg, M., Hummel, W., Wandrey, C. and M€ uller, M. (2001) Chemical Communications, 57–58. (k) Kometani, T., Sakai, Y., Matsumae, H., Shibatani, T. and Matsuno, R. (1997) Journal of Fermentation and Bioengineering, 84, 195–199. (l) Dehli, J.R. and Gotor, V. (2001) Tetrahedron: Asymmetry, 12, 1485–1492. 30 (a) Nakamura, K., Miyai, T., Ushio, K., Oka, S. and Ohno, A. (1988) Bulletin of the Chemical Society of Japan, 61, 2089–2093. (b) Z€ uger, M.F., Giovannini, F. and Seebach, D. (1983) Angewandte ChemieInternational Edition, 22, 1012. (c) Seebach, D., Z€ uger, M.F., Giovannini, F., Sonnleitner, B. and Fiechter, A. (1984) Angewandte Chemie-International Edition, 23, 151. (d) Matzinger, P.K. and Leuenberger, H.G.W. (1985) Applied Microbiology and Biotechnology, 22, 208–210. 31 (a) Hasegawa, J., Ogura, M., Tsuda, S., Maemoto, S., Kutsuki, H. and Ohashi, T. (1990) Agricultural and Biological Chemistry, 54, 1819–1827. (b) Nakamura, K., Inoue, Y., Matsuda, T. and Ohno, A. (1995) Tetrahedron Letters, 36, 6263–6266. (c) Nakamura, K., Fujii, M. and Ida, Y. (2001) Tetrahedron: Asymmetry, 12, 3147–3153. (d) Allan, G.R. and Carnell, A.J. (2001) Journal of Organic Chemistry, 66, 6495–6497. (e) Takemoto, M. and Achiwa, K. (1998) Phytochemistry, 49, 1627–1629.

j229

9 Biooxidations in Chiral Synthesis Marko D. Mihovilovic and Dario A. Bianchi

9.1 Introduction

Enzyme-mediated oxidation reactions offer highly diverse options for the modiﬁcation of existing functional groups as well as for the introduction of novel function in chiral catalysis. Biooxidations often enable us to obtain complementary solutions to metal-assisted transformations and organocatalysis and are considered one of the important strategies of green chemistry. The biooxidations particularly discussed within this chapter focus on enantioselective transformations yielding chiral building blocks for subsequent elaboration. The exceptional chemoselectivity in the majority of biological oxidation reactions enables highly atom efﬁcient synthetic routes by limiting the utilization of protecting groups to a minimum, and hence signiﬁcantly shortening the number of steps in an elaborate sequence toward complex target compounds. Combined with the usually high regioselectivity of enzyme-mediated reactions and the frequently encountered considerable promiscuity of enzymes vis-a-vis nonnatural substrates, such biooxidations are often highly predictable and allow for their utilization in the development of synthetic strategies. In particular, the utilization of molecular oxygen by oxygenases as primary oxidant in combination with mild reaction conditions makes them very appealing catalytic entities also for industrial applications fulﬁlling high safety standards. The application of biooxidation reactions both in synthetic laboratories and in ﬁne chemical industry is hampered to a certain degree by the cofactor dependence of these enzymes as a general challenge in redox biocatalysis. Careful process optimization is required to be implemented in up-scaling efforts [1]. Biotransformations utilizing isolated proteins usually require nicotinamide cofactors (reduced nicotinamide adenine dinucleotide (NADH), reduced nicotinamide adenine dinucleotide phosphate (NADPH), etc.), which have to be recycled in order to enable an economically reasonable process. Hence, an auxiliary substrate has to be added in order to regenerate the cofactor in the required oxidation state ultimately closing the catalytic cycle. In general, this complicates the application in synthetic chemistry by increasing

Asymmetric Organic Synthesis with Enzymes. Edited by Vicente Gotor, Ignacio Alfonso, and Eduardo Garcıa-Urdiales Copyright Ó 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 978-3-527-31825-4

j 9 Biooxidations in Chiral Synthesis

230

the preparative efforts; a more complicated experimental setup is required and product isolation is more demanding. This problem is often solved by creating a closed-loop system with an additional enzyme for cofactor regeneration [2]. Suitable systems are available for recycling of the most abundant cofactors NADH and NADPH as well as their oxidized forms. Recent progress in this ﬁeld focused on the utilization of cheap auxiliary substrates like formate, glucose, or phosphite in combination with promiscuous systems for the recycling of phosphorylated and nonphosphorylated nicotinamide systems. Such systems bear the advantage of being orthogonal to the particular conversion of substrate to product, thus minimizing the danger of unwanted side reactions in the context of cofactor regeneration. Several strategies for nonenzymatic coenzyme recycling have been proposed [3]. Attractive concepts for industrial applications include, in particular, electrochemical cofactor recycling [4] as well as the application of metal complexes (especially Rh systems) to take over the role of the dehydrogenase in closed-loop systems [5]. Another option for recovering cofactors in sufﬁcient amounts is represented by the application of living whole-cells [6,7]. While the microorganism remains intact and its metabolism is operational, cofactor recycling is taken care of by the cell. However, wild-type strains contain a multitude of additional enzymes that may interfere with the required biotransformation, either resulting in a decrease in yield or stereoselectivity. Such obstacles can be overcome by genetically engineering cells to overproduce the required biocatalyst. As a consequence, the desired catalytic entity becomes the dominant fraction in the cells proteome and side reactions of competing enzymes become negligible. Eventually, the production of a competing protein can be suppressed by gene knockout to completely delete unwanted biotransformations. Both strategies can be combined if required. Hence, recombinant microorganisms represent efﬁcient biocatalytic entities to produce the required enzyme and simultaneously take care of any cofactor recycling. Redox biocatalysts are often composed of multifunctional domains or represent protein complexes or membrane bound/attached enzymes. Consequently, their puriﬁcation process may become elaborate and their stability is limited. The latter obstacle can be overcome by recent progress in molecular biology to enable knowledge-based tailoring of enzymes for improving their thermal stability without compromising catalytic efﬁciency [8]. Application of whole-cells again may be an appealing option as it avoids the protein puriﬁcation process, the biocatalysts remain in their native environment, and no disruption of the cellular compartments is required. In biooxidation reactions, a majority of transformations is still performed utilizing living whole-cells or a variation thereof (lyophilized cells, crude cell extracts, etc.) to minimize puriﬁcation efforts by concomitant facile cofactor recycling and maximum biocatalyst activity. The literature describing the characterization and exploitation of a large diversity of biooxidation catalysts has become very rich in recent years and is comprehensively covered by books and reviews in journals [9–15]. Consequently, the current chapter sets out to highlight a few representative and recent examples covering major developments in the ﬁeld since the beginning of this millennium. The major focus

9.2 Oxidations of Alcohols and Amines

is on contributions of particular relevance for synthetic elaboration. With respect to reaction types, the area of stereoselective biooxidations can be roughly subdivided into four major topics: oxidation reactions (oxidation of alcohols and amines) monooxygenation reactions (hydroxylations, epoxidations, Baeyer–Villiger oxidations, and heteroatom oxidations) dioxygenation reactions (aryl dihydroxylation, peroxidation, and alkene cleavage) halogenation reactions. Several reaction types and functional group transformations will be outlined in the following sections with a major emphasis on those biocatalytic processes of major impact on enantioselective synthesis and chiral product preparation.

9.2 Oxidations of Alcohols and Amines

The enzymatic oxidation of alcohols and amines is catalyzed by various oxidoreductases of the dehydrogenase, oxidase, or peroxidase type [16–19]. This transformation may result in the destruction of chirality (at least at the center of reaction), as is the case in the comprehensive conversion of secondary alcohols to carbonyl species. The unspeciﬁc biotransformation can be utilized for cofactor regeneration either in a coupled enzyme or in a coupled substrate approach; in the latter case, the oxidizing enzyme also acts as a stereoselective dehydrogenase. The regioselective oxidation of a particular hydroxyl group may be exploited in selected cases of multifunctional substrates. Certainly, kinetic resolutions or desymmetrizations of bifunctional precursors can be utilized for the production of optically active compounds. The incorporation of a biooxidation step into a redox tandem sequence may be of particular interest, as such a strategy results in a dynamic resolution process, ultimately converting a racemic substrate into a homochiral product. Such transformations are particularly valuable as they overcome the 50% limitation of classical resolutions. Both isolated enzymes and whole-cell systems have been successfully applied in this context. Table 9.1 provides an overview of particularly relevant biocatalytic systems for the biooxidation of alcohols and amines. In the subsequent sections representative examples will be outlined, applying oxidation reactions for the conversion of CO and CN groups. 9.2.1 Regioselective Oxidation of Alcohols

An impressive indication of the high regioselectivity of hydroxysteroid dehydrogenases (HSHDs) was reported for the oxidation of various hydroxyl groups at the steroid core of bile acids [26] (Scheme 9.1). The hydroxy-substituents at positions 3, 7, and 12 could be selectively addressed depending on the hydroxysteroid

j231

j 9 Biooxidations in Chiral Synthesis

232

Table 9.1 Representative isolated enzymes for the oxidation of alcohols.

Representative substrate profile

Biocatalyst/origin Horse liver alcohol dehydrogenase (HLADH) Yeast alcohol dehydrogenase (YADH) Saccharomyces cerevisiae Alcohol dehydrogenase (TBADH) from Thermoanaerobium brockii Glycerol dehydrogenase (GDH) Schizosaccahromyces pombe, Cellulomonas Hydroxysteroid dehydrogenases (HSDH) Arthrobacter BP2 ADH from Lactobacillus keﬁr ADH from Rhodococcus erythropolis Amino acid oxidase (L-AAO) from Proteus myxofaciens Amino acid oxidase (D-AAO) from porcine kidney Monoamine oxidase (MAO-N) from Aspergillus niger and mutants

[20] [21] [20] [22] [9,19] [9,19] [9,19] [23] [24] [25]

dehydrogenases (HSDH) utilized [27]. In addition to regioselectivity, the absolute stereochemistry was also differentiated by suitable biocatalysts. As the direction of biotransformations could be inﬂuenced upon addition of auxiliary substrates (formate versus pyruvate), stereoinversion of particular hydroxyl groups was realized upon sequential biooxidation with a- and b-HSHDs, and vice versa [28]. This technology platform was utilized in the preparation of Gd-labeled cholanoic acids for magnetic resonance imaging [29]. A similar platform of enzymes is available for the regioselective oxidation of carbohydrates (Scheme 9.1) [30]. Carbohydrate dehydrogenases like glucose dehydrogenase (GDH) are capable of speciﬁcally attacking the sugar at the anomeric center (position 1) leading to gluconolactone, which is ultimately hydrolyzed to gluconic acid; this reaction is often utilized in cofactor recycling systems [31]. Pyranose oxidases (P2Os) of usually fungal origin oxidize the 2-position in various pyranose derivatives (mono- and disaccharides) [32], and can be utilized in a chemoenzymatic

12-HSDH GAOX OH

GDH

R HO

O

HO HO 3-HSDH

OH OH

Scheme 9.1 Regioselective alcohol oxidations.

OH OH

7-HSDH

OH

P2O

9.2 Oxidations of Alcohols and Amines

j233

sequence (Cetus process) to convert D-glucose to D-fructose [33,34]. Galactose oxidase (GAOX) can be utilized for the speciﬁc conversion of the 6-hydroxyl to aldehydes or carboxylic acids [35,36]. Applications of such regioselective biooxidations in the modiﬁcation of sugar derivatives are particularly important to access food additives. For example, biotransformation of D-galactitol with whole-cells of Enterobacter agglomerans gives the low-calorie sweetener tagatose in 86% chemical yield [37]. In addition, such siteselective biotransformations were utilized in the laccase-mediated [38] derivatization of bioactive compounds like asiaticoside with site-selective oxidation of a particular 6-hydroxyl in the glycon moiety of the target [39].

9.2.2 Desymmetrizations of Diols

The desymmetrization of meso-diols to chiral lactones undoubtedly represents the most prominent reaction by horse liver alcohol dehydrogenase (HLADH) in asymmetric synthesis [40,41]. This enzyme is particularly promiscuous to accept a large range of structurally diverse diols, and even strained ring systems [42] are readily converted to lactones in high chemical yields and optical purity (Scheme 9.2). Multiple chiral centers can be established in a single operation on preparative scale with excellent stereoselectivity (usually >99% ee) and the enzyme is highly selective for the biotransformation of S-alcohols with other oxidizable functional groups (e.g. oleﬁns, sulfur) remaining intact. The overall conversion consists of two biocatalytic steps. Within the ﬁrst biooxidation, a chiral hydroxy-aldehyde is formed, which is in equilibrium with a lactol species. This compound is then a substrate for the second oxidation. A very simple and efﬁcient cofactor recycling strategy can be set up by adding FMN for the regeneration of NADþ and a facile protocol was established to access multigram quantities. Various biocatalytic options have been presented for the desymmetrization of meso-diols to chiral hydroxyl-ketones. A particularly facile system is represented by

OH

OH HLADH OH

O

HLADH O

CHO OH

NAD+

NADH

NAD+

FMN

FMNH2

Scheme 9.2 Desymmetrization of meso-diols by HLADH.

O NADH

j 9 Biooxidations in Chiral Synthesis

234

Biooxidation

OH

O

R.ruber

( )n meso

S

OH

OH

( )n OH O

Bioreduction

O R

R.ruber

( )n

( )n

O

OH

O

OH R

( )n

R

rac O

OH

OH

R.ruber

OH

R

O

( )n

O OH

+

R

( )n

OH

OH

Scheme 9.3 Biooxidation of diols by lyophilized cells of R. ruber.

lyophilized cells of Rhodococcus ruber DSM 44541, as this catalyst can operate bidirectionally depending on the auxiliary substrate added for cofactor recycling. Bioreductions of diketones can be performed upon concomitant oxidation of 2-propanol to acetone yielding S-hydroxyketones, whereas biooxidations of diols are possible in the presence of excess acetone giving R-hydroxyketones (Scheme 9.3) [43]. A successful application of this biocatalyst among several others was reported for the desymmetrization of terminal secondary diols. Substrates bearing o-1 dihydroxyls, in particular, are converted efﬁciently and in good stereoselectivity (depending on the chain length) to hydroxyketones. The biocatalyst was also used for the resolution of diols incorporating a primary hydroxyl function; in this case, the catalytic species is highly selective for the biooxidation of the secondary alcohol, with the primary OHgroup remaining intact. 9.2.3 Kinetic Resolution of Primary and Secondary Alcohols

An interesting application of kinetic resolutions of alcohols is reported for the synthesis of raspberry ketone using lyophilized cells of various Rhodococcus sp. in the presence of acetone as hydrogen acceptor in a hydrogen transferlike process (Scheme 9.4) [44]. In this particular application, the prime interest was in the nonchiral ketone product (and not the optically enriched R-alcohol), as this biotransformation allows for a sustainable oxidation and may qualify as the methodology to access a product of biological origin in a production process. Similarly, a stereoisomeric mixture of carveol was resolved by Rhodococcus erythropolis DCL14 to ()-carvone and ()-cis-carveol, another important fragrance [45]. In case of primary alcohol substrates, biooxidation can also proceed to the carboxylic acid, enabling a facile separation of the chiral products by simple extraction. Whole-cells of Gluconobacter oxydans were utilized to produce S-2-phenylpropanoic acid and R-2-phenylpropionic alcohol in excellent yields and optical purities (Scheme 9.4) [46].

9.2 Oxidations of Alcohols and Amines OH

j235

O

OH

R. ruber DSM 44541

+

10% acetone HO

HO

HO Raspberry ketone

OH

S COOH

G. oxydans

HO

HO

R

+ HO

Scheme 9.4 Kinetic resolution by alcohol oxidation toward chiral products.

9.2.4 Deracemization of Secondary Alcohols

Biooxidative deracemization of racemic sec-alcohols to single enantiomers [47,48] is complementary to combined metal-assisted lipase-mediated strategies [49,50]. In general, deracemization can be realized by either an enantioconvergent, a dynamic kinetic resolution, or a stereoinversion process. The latter concept is particularly appealing, as only half of the substrate needs to be converted, as the remaining half already represents the product with correct stereochemistry. A two-step process can be set up by combining a biooxidative kinetic resolution with a stereoselective bioreduction, as has been illustrated with two dehydrogenases originating from R. erythropolis and Lactobacillus keﬁr (Scheme 9.5) [51]. Again, both enzymes are operational in both directions, allowing biooxidations and bioreductions depending on the cofactor recycling options. The two enzymes are complementary with respect to substrate acceptance and stereopreference. Consequently, both antipodal chiral alcohols are accessible depending on the order of usage of the two enzymes. Upon kinetic resolution with the S-speciﬁc NADH-dependent alcohol dehydrogenase (ADH) from R. erythropolis (ADHRhodo), S-alcohol is oxidized in a kinetic resolution to ketone and R-alcohol remains unconverted. When the second transformation is then carried out with R-speciﬁc and NADPH-dependent biocatalyst from L. keﬁr (ADHLacto) under reduction conditions, the ketone in the mixture of the ﬁrst conversion is reacted in a convergent step to R-alcohol only. Inverted order of steps under proper mode for the cofactor recycling gives access to the corresponding S-product. A similar set of enzymes was described for the complementary deracemization of a-hydroxyacids [52,53]. Such isolated enzyme approaches for deracemization have a clear disadvantage in that they require two operational manipulations with an intermediate recovery step. A one-pot strategy is offered by employing whole-cell biotransformations with strains containing set(s) of complementary dehydrogenases operating in both biooxidative and bioreductive modes. Trace amounts of the intermediate ketone species can be isolated in several cases. In order to lead to an efﬁcient deracemization

OH

j 9 Biooxidations in Chiral Synthesis

236

O

OH

Biotransformation 1

+

ADHRhodo

R

ADHLacto

OH

Biotransformation 2 O NAD(P)+ NAD(P)H

OH

ADHLacto

+

NADH-oxidase

O2

S

ADHRhodo H2O2

H2O

Catalase Scheme 9.5 A two step process for the deracemization of sec-alcohols by two dehydrogenases.

process, the dehydrogenases involved have to possess antipodal stereopreference and the ultimate reduction step needs to be irreversible. An alternative effect was reported for deracemizations with Cunninghamella echinulata [54]: in this case, both oxidative and reductive biotransformations commence on the S-enantiomer. However, the biooxidation progresses with superior stereoselectivity compared to the bioreduction, thus leading to a gradual increase in R-alcohol (Scheme 9.6). Several suitable whole-cell systems have been identiﬁed for deracemization biotransformations on a large diversity of substrates, as compiled recently [48]. In particular, heterocyclic alcohols were successfully converted by Sphingomonas [55]. Access to enantiocomplementary products was achieved with various strains of Aspergillus [56] or Rhizopus [57]. Biotransformations can even be accomplished with yacon and ginger [58]. Substrate titers were reported up to 8 g l1 for Candida parapsilosis mediated biotransformations [59].

OH

OH

+

Ph NHBz

O

Highly selective biooxidation of

Ph NHBz

S-alcohol

Partially selective bioreduction to S-alcohol Scheme 9.6 Deracemization by Cunninghamella echinulata by imperfect enzymatic interconversion.

OH

+

Ph NHBz

Ph NHBz

9.3 Oxygenation of Nonactivated Carbon Centers

NH2 R'

S

R''

+

NH2 R'

R

R''

NH

Biocatalytic step: Highly selective biooxidation of S-amine

R'

R''

+

NH2 R'

R

R''

Unselective chemical reduction (NH3.BH3) of imine to racemic amines Scheme 9.7 Deracemization of amines in a chemoenzymatic process.

9.2.5 Deracemization of Amines

In light of the paramount role of nitrogen-containing products, transfer of the aboveoutlined methodology for deracemization to amines represents a major progress [60]. A major accomplishment in the ﬁeld was the identiﬁcation of a nonstereospeciﬁc reduction methodology for the conversion of imines to amines, which operates under conditions of isolated enzyme biocatalysis for a stereoselective oxidation of amine to imine (Scheme 9.7). In particular, boranes are sufﬁciently water stable to enable such an integrated process similar to a heterogeneous variation using Pd/C and HCOONH4 [61]. Biocatalysts to access enantiocomplementary products are available from the group of amino acid oxidases (AAOs): while L-AAO from Proteus myxofaciens selectively oxidizes L-amine precursors [61], subsequently enabling access to optically pure D-amines, the antipodal product is available upon application of commercially available porcine kidney D-AAO [62]. The methodology has been successfully extended to the interconversion of a range of nonnatural diastereoisomeric a-amino acids [63,64]. Upon mutagenesis of the monoamine oxidase from Aspergillus niger (MAO-N) within several rounds of directed evolution [65], variant biocatalysts were identiﬁed with largely expanded substrate acceptance, enabling also the deracemization of tertiary amines incorporating straight-chain and cyclic structural motifs [66].

9.3 Oxygenation of Nonactivated Carbon Centers

The functionalization of nonactivated carbon centers in natural and synthetic compounds represents one of the most powerful conversions enabled by biocatalysis, in particular as it is also highly regio- and stereoselective [12]. This biotransformation is a major pathway in metabolic degradation of xenobiotics in order to increase their hydrophilicity for subsequent excretion from the biological system. The investigation of metabolites of drugs require speciﬁc access to such biohydroxylated compounds and several microbial systems mimicking biocatalytic properties of human tissue have been utilized in this context [67]. In addition, several bioactive compounds are

j237

j 9 Biooxidations in Chiral Synthesis

238

Catharanthus roseus 3

R 2

R 1

4

Rhodococcus opacus O

R

Sphingomonas

R

( )n

( )m N PG

R1: =O, -OH R2: H, Me

PG: COOPh, COOBn n, m = 0, 1

Scheme 9.8 Biohydroxylation sites at representative scaffolds (filled arrows: major sites, open arrows: minor sites).

synthesized industrially involving microbial hydroxylation such as hydrocortisone or the cholesterol regulating drug pravastatin [14]. 9.3.1 Hydroxylations by Wild-Type Whole-Cells

A multitude of microorganisms have been reported to possess hydroxylating abilities on a large diversity of organic scaffolds [68] as outlined in Scheme 9.8 on selected examples; both a- and b-alcohols for essentially all positions can be accessed by microbial hydroxylation at the steroid ABCD-ring system [69]. Terpenes can be oxidized with high regio- and stereoselectivity in axial and equatorial positions [70,71] and the hydroxylation at primary or secondary allylic positions represents a particularly interesting aspect from a synthetic chemists point of view [72,73]. Benzylic positions can be biooxidized by Bacillus sp. in moderate to excellent enantioselectivity [74,75]. Heterocyclic systems can be efﬁciently hydroxylated by a Sphingomonas sp. and the stereoselectivity of this biooxygenation (for n 6¼ m) can be inﬂuenced by the nature of the nitrogen protecting group PG to give enantiocomplementary products [76,77]. Such a strong inﬂuence on the stereoselectivity of enzymatic hydroxylation reactions by minor modiﬁcations to the substrate have been exploited in the docking/protecting (d/b) group concept [78]. The strategy was initially employed in Beauveria bassiana mediated hydroxylation reactions of cycloketone precursors. As the wild-type strain displayed predominant reduction of the carbonyl center, proper modiﬁcation of the ketone signiﬁcantly improved the yields on oxygenation products. Hence, this d/b-group facilitated the biooxygenation and concomitantly prevented unwanted side reactions [79]. Structural modiﬁcation of the d/b-group also improved the stereoselectivity of the biotransformation (Scheme 9.9: R¼H, 40% ee; R ¼ t-Bu, 89% ee), in particular, when applying chiral auxiliaries [80]. Inversion of chirality at

9.3 Oxygenation of Nonactivated Carbon Centers R

NBz

O

B. basseana

R

OH

NBz

j239

OBn O

O

O COOH

COOH

N

N

C. blakesleeana

O

O

OH

HO

Scheme 9.9 Biohydroxylation utilizing the docking/protecting group concept.

the d/b-group led to antipodal products, though in low optical purity. A similar concept was utilized for the oxygenation of aldehydes using the oxazolidine moiety, carboxylates utilizing the benzoxazole d/b-group, and alcohols via isosaccharine derivatives [81]. The nature of the nitrogen protecting group also played a signiﬁcant role in the chemoenzymatic total synthesis of epibatidine, which shall be outlined as an example for the synthetic elaboration of the regioselective biooxidation product of a nonnatural precursor. B. bassiana mediated hydroxylation of the aza-norbornane system enabled functionalization for the subsequent introduction of the pyridine system (Scheme 9.10) [82,83].

NPG

B. bassiana

NPG

NPG

NPG

N

Cl

O OH

Epibatidine

Scheme 9.10 Chemoenzymatic access to epibatidine by biohydroxylation.

9.3.2 Hydroxylations by Recombinant Cytochrome P450 Monooxygenases

Cytochrome P450 monooxygenases (CytP450MOs) have received increasing attention in the context of the hydroxylation of aliphatic carbon centers [84–86]. Usually, (recombinant) whole-cell biotransformations are utilized, but isolated proteins in a biphasic environment have also been studied [87]. Such enzymes usually represent a three-component protein complex consisting of a cytochrome P450 reductase, an iron–sulfur electron transfer protein, and the actual cytochrome P450 monoxygenase, which acts as an electron relay to activate molecular oxygen as primary oxidant. A widely utilized prototype enzyme of this group is P450cam from Pseudomonas putida, for which the structure was solved by X-ray diffraction [88]. The wild-type enzyme catalyzes the hydroxylation of camphor to incorporate a 5-exo-hydroxyl, and based on the knowledge of the enzyme conﬁguration, diverse transformations were enabled by site-directed mutagenesis [89–91]. In this context, a strategy for the generation of large plasmid libraries of mutants of this enzyme was presented [92].

j 9 Biooxidations in Chiral Synthesis

240

In contrast, the soluble CytP450 from Bacillus megaterium (P450 BM-3) is catalytically self-sufﬁcient as it contains only the MO and dehydrogenase domains within a single polypeptide chain. Although the wild-type enzyme has a rather limited substrate proﬁle to hydroxylate fatty acid chains at the subterminal position, the catalytic repertoire of the enzyme was successfully extended by directed evolution [93,94]. Together with P450 enzymes of fungal origin, this enzyme has received particular attention for the regioselective hydroxylation of various positions in fatty acids [95,96]. The complementary approach to modify the stereoselectivity of such enzymes by substrate engineering was successfully demonstrated using the d/p-group concept and excellent diastereo- and enantioselectivities were obtained with mutant P450 BM-3 in cases of insufﬁcient selectivity of the wild-type enzyme [97]. Recombinant whole-cells expressing P450 enzymes were successfully applied in natural product modiﬁcation, in particular with terpene substrates (Scheme 9.11). A representative of this protein superfamily originating from Thermus thermophilus displayed regioselective hydroxylation of b-carotene to zeaxanthin [98]. a-Pinene is a waste product of the wood industry and was converted to fragrance compounds cis-verbenol (82% selectivity) and verbenone (32% selectivity) by various mutants of P450Cam [99]. b-Ionone is biooxidized regioselectively at the allylic position by wild-type and mutant P450 BM-3 [100]. A comparative study of P450Cam and BM-3 was conducted for the biooxygenation of valencene to nootkatol with the prior

)2

)2 HO

β-carotene

Zeaxanthin

+ OH (+)-cis-verbenol

(+)-α-pinene O

β-ionone

O

OH

HO

4-hydroxyβ-ionone

(+)-trans-nootkatol

(+)-valencene

HO

(−)-trans-carveol

O (+)-verbenone

OH

(−)-lemonene

(+)-lemonene

Scheme 9.11 Terpene hydroxylation by recombinant P450s.

(+)-cis-carveol

9.4 Enzymatic Epoxidation

enzyme displaying superior chemo- and regioselectivity [101]. An interesting enantiodivergent behavior of lemonene-P450 from Mentha sp. was reported: while biooxidation of ()-lemonene gave ()-trans-carveol, selectively, the antipodal substrate gave the (þ)-cis-diastereomer (accompanied by minor amounts of other oxygenated products) [102]. 9.3.3 Hydroxylation via Hydroperoxide Formation

Applications of peroxide formation are underrepresented in chiral synthetic chemistry, most likely owing to the limited stability of such intermediates. Lipoxygenases, as prototype biocatalysts for such reactions, display rather limited substrate speciﬁcity. However, interesting functionalizations at allylic positions of unsaturated fatty acids can be realized in high regio- and stereoselectivity, when the enzymatic oxidation is coupled to a chemical or enzymatic reduction process. While early work focused on derivatives of arachidonic acid chemical modiﬁcations to the carboxylate moiety are possible, provided that a sufﬁciently hydrophilic functionality remained. By means of this strategy, chiral diendiols are accessible after hydroperoxide reduction (Scheme 9.12) [103,104]. OCOR' ( )n

SBLO R

OCOR' ( )n

OH

Reduction R OOH

Hydrolysis

( )n

R OH

Scheme 9.12 Soybean lipoxygenase (SBLO)-mediated oxygenation of fatty acid derivatives.

9.4 Enzymatic Epoxidation

The enzymatic oxygenation of oleﬁns to chiral epoxides is certainly one type of biotransformation that has encountered substantial competition from metal-assisted reactions [105]. Several classical protocols for alkene epoxidation and dihydroxylation have become well-established methods both on laboratory and industrial scale and this technology was recently recognized and awarded a Nobel prize [106]. While biocatalytic epoxidation was hampered by several obstacles (biocatalyst efﬁciency, limited epoxide stability under biotransformation conditions owing to chemical and enzymatic degradation, etc) in the early days, it has also become a valuable tool in organic synthesis. In particular, introduction of recombinant expression systems to improve accessibility of the catalytic entity [107] together with alternative cofactor recycling techniques [108] and optimized fermenter technology [109] allow for robust multigram to pilot-scale bioepoxidations [110]. In the following text, a brief overview of functionalization of various oleﬁns is provided to give an outline of the current synthetic potential of this biooxygenation.

j241

j 9 Biooxidations in Chiral Synthesis

242

R''' O

R''' Steroid or xylane

R''

Monooxygenase

R'

R''

R'

Scheme 9.13 Epoxidation of vinylic alkenes.

Some of the major enzyme groups that facilitate this transformation are hemecontaining MOs of the cytochrome P450 type [111], alkane hydroxylases, xylene monooxygenases, styrene monooxygenases [105], and haloperoxidases [112]. Styrene was successfully oxidized to the S-product both by xylene monooxygenase from P. putida mt-2 [113] and styrene monooxygenase from Pseudomonas sp. VLB120 [114] (Scheme 9.13), with the latter enzyme displaying a particularly large substrate tolerance with excellent stereoselectivity (>99% ee). In this context it is interesting to note that both xylene monooxygenase as well as chloroperoxidase are very selective for mono-epoxidation in case of presence of multiple alkene functionalities [115]. The alkane hydroxylase from Pseudomonas oleovorans is particularly suitable for the epoxidation of terminal aliphatic double bonds and enables rapid access to the bblocker metoprolol (Scheme 9.14) [113,116]. Complementing this regioselectivity, chloroperoxidases are particularly suitable biocatalysts for the epoxidation of (cis substituted) subterminal oleﬁns [112,117]. This enzyme also accepts terminal oleﬁns and is utilized for the efﬁcient synthesis of R-mevalono-lactone [118]. The alkane-degrading bacterium Shingomonas sp. HXN-200 was identiﬁed as a particularly rich source of oxygenating enzymes. An alkane monooxygenase from this organism was used for the functionalization of various nitrogen-containing heterocycles. Combination of enzyme-mediated monooxygenation with chemical epoxide hydrolysis gave access to trans-diols (Scheme 9.15) [119]. This transformation

P. oleovorans O O

O

P. oleovorans

R

O

NH-iPr R Metoprolol (R = CH2CH2OMe)

R

R

OH O

Chloroperoxidase H2O2

R O HO

Chloroperoxidase COOEt

COOEt H2O2

O

O

Mevalonolactone

Scheme 9.14 Epoxidation of aliphatic alkenes and applications in bioactive compound synthesis.

9.5 Baeyer–Villiger Oxidations

O

OH OH

Shingomonas N

N

N

PG

PG

PG

Scheme 9.15 Epoxidation of heterocyclic systems to trans-diols.

serves as a representative example for chemoenzymatic reaction sequences to trans-dihydroxyls involving an epoxide intermediate. It should also be mentioned that chemoenzymatic epoxidation reactions have been proposed using lipase and H2O2 on multigram scale [120]. Another interesting biooxygenation reaction with alkenes, recently identiﬁed, represents an enzymatic equivalent to an ozonolysis. While only studied on nonchiral molecules, so far, this cleavage of an alkene into two aldehydes under scores the diversity of functional group interconversions possible by enzymatic processes [121,122].

9.5 Baeyer–Villiger Oxidations

The enzyme-mediated Baeyer–Villiger oxidation to chiral lactone intermediates has received considerable attention in recent years as it offers several advantages in chemo-, regio-, and stereoselectivity compared to other catalytic strategies [123]. Flavin-containing Baeyer–Villiger monooxygenases (BVMOs) represent natures equivalent of conventional peracids or de novo designed metal complexes [124]. These enzymes display a remarkably broad acceptance proﬁle for nonnatural substrates [125–127]. The mechanism of the enzymatic reaction is well established and involves an FAD peroxy anion as key reactive species, which undergoes nucleophilic addition at a carbonyl group of the substrate toward the classical Crigee intermediate with concomitant rearrangement to an ester or lactone [128,129]. In addition, the 3D structure of phenylacetone monooxygenase (PAMO [130]) was successfully established by X-ray diffraction as the ﬁrst representative of this type of ﬂavoproteins [131]. This contribution provided valuable insights into the molecular architecture of these enzymes and conformational changes throughout the catalytic cycle. This structure also serves as a template for models of other BVMOs. On the basis of the importance of chiral lactones in asymmetric synthesis, Baeyer– Villiger biooxidations of cyclic ketones have received by far the most attention. Cyclohexanone monooxygenase from Acinetobacter NCIMB 9871 (CHMOAcineto) plays a paramount role in biocatalytic applications of BVMOs and the majority of typical transformations were identiﬁed using this enzyme [132]. However, a large diversity of different BVMOs have become available in recent years. Various techniques in molecular biology were used to access novel genes, either by knowledge-based [133] or random strategies [134]. Studies to investigate their potential for biocatalytic applications in particular utilized recombinant whole-cell expression systems based

j243

j 9 Biooxidations in Chiral Synthesis

244

on Saccharomyces cerevisiae [135] and Escherichia coli [136]. A valuable contribution was the discovery of cyclopentanone monooxygenase from Comamonas NCIMB 9872 (CPMOComa), cyclododecanone monooxygenase (CDMO), and cyclopentadecanone monooxygenase (CPDMO) as enzymes accepting large and structurally demanding substrates, ultimately increasing the complexity and number of stereogenic centers of the biooxidation products [137,138]. Table 9.2 provides an overview of BVMOs of relevance for synthetic applications. BVMOs are known to display limited efﬁciency owing to substrate and product inhibition, complemented by possible toxicity in whole-cell biotransformations. However, several studies have outlined strategies to circumvent this obstacle. The particularly appealing approaches involve utilization of nongrowing cells [155] or two-phase fermentation protocols. Applying the substrate feeding–product removal (SFPR) concept, administration of a resin to act as reservoir for both substrate and product has been successfully implemented in recombinant whole-cell biotransformations on various substrates and expression strains [156,157]. This strategy enables access to kilogram-scale biotransformations upon further optimization of the fermentation parameters and equipment [158].

Table 9.2 Recombinant BVMOs of relevance in synthetic applications.

Year BVMO/origin

Identified

Cloned

Representative substrate profile

Aliphatic open-chain monooxygenase (AOCMO) Pseudomonas ﬂuorescens DSM 50106 Baeyer–Villiger monooxygenase (BVMOMtb5) Mycobacterium tuberculosis H37Rv Cyclododecanone monooxygenase (CDMO) Rhodococcus SC1 Cyclohexanone monooxygenase (CHMOAcineto) Acinetobacter NCIMB 9871 Cyclohexanone monooxygenase (CHMOArthro) Arthrobacter BP2 Cyclohexanone monooxygenase (CHMOBrachy) Brachymonas petroleovorans Cyclohexanone monooxygenase (CHMOBrevi1&2) Brevibacterium HCU Cyclohexanone monooxygenase (CHMORhodo1&2) Rhodococcus Phi1 & Phi2 Cyclopentadecanone monooxygenase (CPDMO) Pseudomonas HI-70 Cyclopentanone monooxygenase (CPMOComa) Comamonas NCIMB 9872 Phenylacetone monooxygenase (PAMO) Thermobiﬁca fusca

2006 [139]

2006 [139]

[139]

2006 [140]

2006 [140]

[140,141]

2001 [142]

2001 [142]

[143]

1976 [144]

1988 [145]

[146]

2003 [147]

2003 [147]

[143,148]

2003 [149]

2003 [149]

[143,148]

2000 [150]

2000 [150]

[143,151]

2003 [147]

2003 [147]

[143,148]

2006 [137]

2006 [137]

[137]

1976 [152]

2002 [153]

[148,153]

2004 [131]

2005 [130]

[154]

9.5 Baeyer–Villiger Oxidations

9.5.1 Chemoselectivity

Although BVMOs can also be utilized in sulfoxidation reactions (see Chapter 9.6), the Baeyer–Villiger process is usually favored. This is demonstrated by the chemoselective oxidation of the keto functionality in heterocyclic substrates using CHMOAcineto (Scheme 9.16) [159]. Compounds containing sulfur (X ¼ S, R ¼ H) as well as amine nitrogen (X ¼ NMe/N-allyl, R ¼ H) were selectively converted to the corresponding lactones. Whole-cell-mediated biooxidation was less efﬁcient with more polar acyl protected piperidon systems (X ¼ NAc/NCOOMe, R ¼ H). Tetrahydropyranones (X ¼ O, R ¼ H) were readily accepted by BVMOs and corresponding prochiral substrates were converted to lactones in excellent stereoselectivity (>95% ee) up to chain lengths of C3 (R ¼ Me, Et, n-Pr) [160]. The high chemoselectivity for the Baeyer–Villiger process was utilized in the synthetic elaboration of another hetero-bicyclic substrate. The biooxidation only provides the expected unsaturated lactone in a desymmetrization reaction without compromising the oleﬁn functionality. The biotransformation product was then converted to pivotal intermediates for C-nucleosides like showdomycin, tetrahydrofuran natural products like kumausyne, and goniofufurone analogs in subsequent chemical operations (Scheme 9.17) [161]. The regioselectivity of BVMOs upon biooxidation of substrates incorporating more than one carbonyl center has received only limited attention and trends are somewhat ambiguous. In an early contribution to the biooxidation of a cyclohexan1,4-dione by isolated CHMOAcineto, the sterically less hindered ketone was reported to be converted (Scheme 9.18) [162]. In a recent investigation of fused bicyclodiketones, a more differentiated behavior of the same enzyme was observed when utilizing a crude protein preparation from a recombinant overexpression system [163]. CHMOAcineto is selective for the biotransformation of saturated carbonyl functionalities and enones are not accepted. The enzyme displays preferred conversion of the sterically more demanding carbonyl center in the cis-decaline system (Scheme 9.18, n ¼ 1) within a kinetic resolution process in high optical purity. This is presumably the result of the activity of (an) additional dehydrogenase(s) present in the crude protein preparation reducing the antipodal substrate diketone selectively at the sterically less hindered site. Remarkably, the preference of CHMOAcineto switches in the related cis-hydrindane substrate (n ¼ 0) favoring biooxidation at the less hindered carbonyl center of the cyclohexane core, however, with low conversion. O

O

O

CHMOAcineto R

X

R

R

X

R

X = S, O, N-PG R = H, alkyl Scheme 9.16 Baeyer–Villiger biooxidation of heterocyclic substrates.

j245

j 9 Biooxidations in Chiral Synthesis

246

O NH

O HO

O HO

OH

(+)-Showdomycin O

O O

95% ee

Br

O

O

CPMOComa

O

1S

6S

OAc (+)-Trans-kumausyne Ph

H

O

HO O H Goniofufurone analogs HO

O

Scheme 9.17 Baeyer–Villiger biooxidation of a prochiral heterobicycloketone and subsequent synthetic elaboration in formal total syntheses of natural products.

9.5.2 Desymmetrizations

A major contribution to the ﬁeld in recent years was the identiﬁcation of subclusters of BVMOs with overlapping substrate acceptance providing access to antipodal O

O O

CHMOAcineto

O

O 25%, > 98% ee

O CHMOAcineto O

O 35%, >99% ee

43%, 80% ee

O

O

O S

CHMOAcineto

CHMOAcineto O

R

+

O rac

O

O

O S

S O

O

H 14%

( )n rac

O

H 22%, > 99% ee

Scheme 9.18 Regioselective biooxidation of dicarbonyl substrates.

O

O

R

+ HO

H 32%, > 99% ee

9.5 Baeyer–Villiger Oxidations Table 9.3 Desymmetrization of prochiral cycloketones to enantiocomplementary lactones by CHMO- (CHMOAcineto and CHMOBrevi1) and CPMO-type (CHMOBrevi2 and CPMOComa) enzymes (representative examples). O

H

O Cl

H

O

O

O H

CHMOAcineto CHMOBrevi1 CHMOBrevi2 CPMOComa

H

>99% ee () 95% ee () 60% ee (þ) 48% ee (þ)

92% ee (þ) >99% ee (þ) >99% ee () 99% ee ()

O

5% ee () 71% ee () 94% ee (þ) >99% ee (þ)

lactones [148]. Upon comparison of primary protein sequence, phylogenetic relationship, and biocatalytic performance, a collection of eight bacterial cycloketone accepting BVMOs was classiﬁed into CHMO-type and CPMO-type enzymes with CHMOAcineto and CPMOComa as prototypes for the particular clusters (Table 9.3). This suite of BVMOs is available via whole-cell expression systems and represents a complementary platform of biocatalysts for diverse applications in chiral synthesis. Representatives of this collection were utilized in the enantiodivergent synthesis of the indole alkaloids alloyohimbane and antirhine from a fused bicyclic precursor (Scheme 9.19) [151]. A combined photochemical and biocatalytic approach provided access to bicyclo [4.2.0]octanes via a novel strategy. The Cu-catalyzed [2 þ 2] photocycloaddition of

(−)-Lactone CHMOBrevi1 71% ee H

H O

H

O

BVMO

H N H

N H H

(−)-Alloyohimbane

O H (+)-Lactone CHMOBrevi2 94% ee

H N H (−)-Antirhine

N H H OH

Scheme 9.19 Desymmetrization to enantiocomplementary lactones as pivotal precursors for two indole alkaloids.

j247

j 9 Biooxidations in Chiral Synthesis

248

H CHMOBrevi1 96% ee H i) hν/Cu2+ OPG

ii) Chem. oxidation

O O

H

O H

H CPMOComa 86% ee

HO

H

O O

H

M eOOC

H

Scheme 9.20 Combined photochemical and biocatalytic approach to bicyclo[4.2.0]octanes.

terminal dioleﬁns is only successful when the double bonds are in close proximity for the cyclization reaction. Hence, a cleavable bridge bearing a keto functionality was introduced, which was then submitted to microbial Baeyer–Villiger oxidations. Again, antipodal lactones were obtained with full control over four stereogenic centers utilizing representatives of the CHMO- and CPMO-groups, and the fused system was obtained after chemical hydrolysis (Scheme 9.20) [164]. Biooxidation of cyclobutanones is a particularly useful transformation, as the corresponding chiral butyrolactones represent highly valuable building blocks for a large variety of natural products as well as bioactive compounds [165]. Complementing the arsenal of wild-type BVMOs, mutagenesis studies have been performed on CHMOAcineto and CPMOComa in order to improve their biocatalytic performance. In a random approach, the poor stereoselectivity of wild-type CHMOAcineto for the biooxidation of 4-hydroxycyclohexanone (9% ee R-lactone) was improved signiﬁcantly and even enantiocomplementary products were obtained within one to two generations (Table 9.4) [166]. A remarkable inﬂuence of the nature of amino acid at position 432 was identiﬁed with respect to stereoselectivity. Further investigation of selected mutants of the library showed, that enantiocomplementary biooxidations were also found with other substrates. In addition, the substrate acceptance could be extended to produce antipodal lactones, which were not obtained with wild-type CHMOAcineto [167]. Applying the evolutionary strategy of restricted CASTing, CPMOComa was modiﬁed in a related study by simultaneous exchange of pairs of amino acids in close proximity to the active site of the enzyme [168]. 9.5.3 Kinetic Resolutions

The classical kinetic resolution of racemic substrate precursors allows only access to a theoretical 50% yield of the chiral lactone product, while the antipodal starting material remains unchanged in enantiomerically pure form. The regioselectivity for the enzymatic oxidation correlates to the chemical reaction with preferred and exclusive migration of the more nucleophilic center (usually the higher substituted a-carbon). The majority of cycloketone converting BVMOs (in particular CHMOAcineto)

9.5 Baeyer–Villiger Oxidations Table 9.4 Mutagenesis of CHMOAcineto toward enantiodivergent Baeyer–Villiger oxidations.

Lactone

O

OH

O

O

BnO

O H

Cl

O O

H

O O

Mutant

Amino acid modificationsa

eeb

Wild type 1-C2-B7 1-E12-B5 1-H7-F4 1-F1-F5 2-D19-E6 1-F4-B9 1-K6-G2 1-K2-F5

— F432A, K500R F432I L426P, A541V L143F L143F, E292G, L435Q, T464A D41N, F505Y K78E, F432S F432S

10% () 34% () 49% () 54% () 40% () 90% () 46% (þ) 78% (þ) 79% (þ)

Wild type 1-K2-F5 1-E12-B5

— F432S F432I

53% (þ) 83% () 66% (þ)

Wild type 1-K2-F5 1-H7-F4

— F432S L426P, A541V

99% () 99% () 60% (þ)

Wild type 1-K2-F5 1-H7-F4

— F432S L426P, A541V

n.c.c 90% () 60% (þ)

a

Position and substitution of amino acids in one-letter coding. Sign of speciﬁc rotation in parentheses. c No conversion. b

accepts cyclopentanone (Scheme 9.21, n ¼ 0) and cyclohexanone (n ¼ 1) derivatives and provides access to S-lactones and R-ketones are recovered. The process gives high enantiomeric ratios E for side chains incorporating more than two carbon atoms. The observation that kinetic resolution of a-substituted cyclohexanones give better results than in the series of corresponding cyclopentanones might be explained by the energetically more pronounced difference of axial and equatorial positions on sixmembered rings compared to ﬁve-ring systems. BVMOs are highly tolerant toward a large diversity of functional groups within the side chain [169–173]. It is noteworthy that CPMOComa is the only BVMO reported to date to convert cyclic enones in a kinetic resolution process [174]. Both chiral lactones and ketones have been utilized in asymmetric synthesis of bioactive compounds like lipoic acid [175] and natural products like various insect pheromones [176]. O

O

O O S

R ( )n

( )n

R

R +

R ( )n

= CH2CH2, CH=CH Scheme 9.21 Kinetic resolution of racemic cycloketones by BVMOs.

j249

j 9 Biooxidations in Chiral Synthesis

250

O

O

O pH 9

S

CHMOAcineto1

R

OBn

OBn

O R

Recomb. cells

OBn

85%, 96% ee Scheme 9.22 Dynamic kinetic resolution with CHMOAcineto.

The most important contribution in the area of kinetic resolution via Baeyer– Villiger oxygenation in recent years was the development of a dynamic process facilitating in situ racemization of the substrate to overcome the limitations of 50% maximum yield. Such a racemization reaction requires to be at least as fast as the actual biooxidation step and conditions have to be selected carefully to avoid unwanted racemization of the lactones obtained. Within biooxidation of a functionalized cyclopentanone by recombinant whole-cells expressing CHMOAcineto, signiﬁcant racemization of the substrate ketone was already observed at pH 7 (Scheme 9.22). Increasing the pH to 9 resulted in sufﬁcient acceleration of substrate racemization, and complete conversion to the R-lactone was achieved in 85% yield without compromising the stereoselectivity of the biooxidation (96% ee) [177]. Such a dynamic kinetic resolution can also be performed utilizing the SFPR concept upon selection of a suitable resin to realize substrate racemization [178]. Although a majority of research activities were dedicated to cycloketone converting BVMOs, the recently discovered novel MOs also enable stereoselective oxidation of noncyclic ketones to esters. An aliphatic open-chain monooxygenase (AOCMO) from Pseudomonas ﬂuorescens DSM 50106 displays stereoselective biooxidation of terminal acyl-groups in proximity to hydroxyls (Scheme 9.23). The biooxidation gives acetic OH

O

OH

AOCMO O

( )n

O +

( )n

OH ( )n

O n=3, 5, 7

90–93% ee OAc HO

O

O

( )n

O AcO

O HO

PAMO

Chiral products

82% ee Scheme 9.23 Kinetic resolution of noncyclic ketones to chiral esters.

9.5 Baeyer–Villiger Oxidations

esters as major products and only trace amounts of the regioisomeric methyl esters were detected. Both chiral esters and ketones can be isolated in high optical purity when suitable conditions of the kinetic resolution process are present [139]. The thermostable enzyme PAMO was the ﬁrst BVMO identiﬁed to oxidize enolizable diketones in acceptable stereoselectivity (82% ee). The R-acetate obtained was hydrolyzed to R-hydroxyphenylacetone as an interesting intermediate for various pharmaceutical compounds (Scheme 9.23) [179]. 9.5.4 Regiodivergent Biooxidations

In the majority of cases, strict migratory preference of the more nucleophilic center is observed for both chemical and enzymatic oxidations based on the difference in electron density between carbons bearing an additional group and those with all hydrogen substituents. However, deviations from this rule of thumb were reported in particular for enzymatic transformations. Some racemic precursors are oxidized by BVMOs into two types of regioisomeric lactones in a resolution process: migration of the more substituted carbon atom generates the expected normal lactone, while abnormal lactone is formed by migration of the less substituted carbon atom (Scheme 9.24). This behavior was initially observed during early studies on CHMOAcineto with fused bicycloketones bearing a cyclobutanone structural motif [180] and a mechanistic rational was proposed [181]. All chiral products as well as enantiomerically enriched substrate ketones from such transformations are valuable building blocks in asymmetric synthesis [182,183]. While CHMO-type enzymes in general display such a behavior, CPMO-type biocatalysts give H

O O

CPMO-type

O

H Normal lactone

CHMO-type

H

H

O O

+

H Normal lactone

rac

BVMOMtb5

H

H Abnormal lactone

O H Abnormal lactone

O O

O

H

O

+ H Chiral ketone

Scheme 9.24 Regiodivergent biooxidation of fused bicycloketone precursors.

j251

j 9 Biooxidations in Chiral Synthesis

252

normal lactones as predominant products, though in racemic form [184]. Nevertheless, this represents an appealing entry to such compounds when starting from optically pure starting material in the case of functional incompatibility with chemical oxidation. In addition, the separation of regioisomeric lactones may be troublesome and can be circumvented in that case. The identiﬁcation of a novel BVMO from Mycobacterium tuberculosis (BVMOMtb5) complements this toolbox, as this particular biocatalyst performs a classical kinetic resolution instead of a regiodivergent oxidation with complete consumption of substrate [140]. Notably, this enzyme accepts only one ketone enantiomer and converts it selectively to the abnormal lactone while the antipodal substrate remains unchanged (Scheme 9.24) [141]. Within monocyclic ketones, a similar inversion of the migratory preference was only reported for the biooxidation of terpenones with CHMOAcineto: while ()-dihydrocarvone is converted to the expected normal lactone, migration of the less substituted carbon center was observed for the biooxidation of antipodal (þ)-dihydrocarvone, ultimately providing the abnormal lactone [185]. A related situation is found in the case of b-substituted cycloketones: here, the electronic difference between the two a-carbons is almost insigniﬁcant, resulting in unselective migration upon chemical oxidation. BVMOs have a particularly different behavior, as they can inﬂuence the stereo- and/or regioselectivity of the biooxidation. In the latter case, the distribution of proximal and distal lactones is affected by directing the oxygen insertion process either into the bond close or remote to the position of the b-substituent. Consequently, a regioisomeric excess (re) can be deﬁned for this biotransformation, similar to enantiomeric excess or diastereomeric excess values [143]. Again, a distinctly different selectivity was reported for CHMO-type enzymes vis-a-vis CPMO-type enzymes. While there is currently no enzymatic system available to allow for the regioselective formation of optically pure products in the series of b-substituted cycloketones, a combined chemical and biocatalytic approach was outlined as a versatile strategy for accessing both proximal and distal lactones [186]: stereoselective reduction of 3-substituted cyclohexenones or cyclopentenones in the presence of a chiral ligand affords enantiomerically pure/enriched ketones, which are then regioselectively oxidized by BVMOs. Again, cases are reported where CHMOAcineto displays regiodivergent oxidations, while CPMOComa usually gives only proximal lactones (Scheme 9.25). O

( )n

Stereoselective reduction R

L*

O

O BVMO ( )n

n=0: L* = (S)-p-tol-BINAP n=1: L* = (S)-BIPHEMP

R

O O

O

+ ( )n

R Proximal lactone

Scheme 9.25 Combined chemo- and biooxidative route to chiral proximal and distal lactones.

( )n R Distal lactone

9.6 Heteroatom Oxidations

9.6 Heteroatom Oxidations

The oxidation of heteroatoms and, in particular, the conversion of sulﬁdes to asymmetric sulfoxides has continued to be a highly active ﬁeld in biocatalysis. In particular, the diverse biotransformations at sulfur have received the majority of attention in the area of enzyme-mediated heteroatom oxidation. This is particularly due to the versatile applicability of sulfoxides as chiral auxiliaries in a variety of transformations coupled with facile protocols for the ultimate removal [187]. Enzyme-mediated chiral sulfoxidation has been reviewed comprehensively in historical context [188–191]. The biotransformation can be mediated by cytochrome P-450 and ﬂavin-dependent MOs, peroxidases, and haloperoxidases. Owing to limited stability and troublesome protein isolation, a majority of biotransformations were reported using whole-cells or crude preparations. In particular, fungi have been identiﬁed as valuable sources of such biocatalysts and the catalytic entities have not been fully identiﬁed in all cases. 9.6.1 Sulfide Oxidation

The enzymatic oxygenation process is of particular value as there is a signiﬁcant difference in the formation rates of sulfoxides and sulfones. The initial conversion of sulﬁde to the optically active sulfoxide by an MO is usually very fast compared to the subsequent oxidation step to sulfone, upon which chirality is lost (Scheme 9.26). In many cases, over-oxidation to sulfone is not observed at all when employing MOs. When a sufﬁcient sterical difference between both substituents at the sulfur center exists, particularly good selectivities are obtained. Hence, aryl–alkyl sulﬁdes were investigated in several comprehensive studies. A particularly valuable contribution to the ﬁeld was the identiﬁcation of sets of oxygenases with overlapping substrate speciﬁcity providing access to antipodal sulfoxides. Such a behavior was observed in whole-cell-mediated biotransformations using Helminthosporium (S-sulfoxides) and Mortierella isabellina (R-sulfoxides) [192]. In the series of haloperoxidases, enzymes from Ascophyllum nodosum were reported to give R-products [193], while S-enantiomers can be accessed using bromoperoxidases from Corallina ofﬁcinalis and Corallina pilulifera [194]. Biotransformations with such enzymes require addition of an oxydans such as hydrogen peroxide or an organic hydroperoxide; in the case of a chiral hydroperoxide species, the stereoselectivity of the biocatalytic system can be

R'

S

MO R''

Sulfide

fast

O S* R' R'' Sulfoxide (chiral)

MO very slow

O R'

S

O R''

Sulfone (nonchiral)

Scheme 9.26 Oxygenation of sulfides to chiral sulfoxides and nonchiral sulfones.

j253

j 9 Biooxidations in Chiral Synthesis

254

affected to a certain extent [195]. A novel microencapsulation strategy may overcome the limited stability of these enzymes [196]. Also, bacterial systems have been applied to simple sulﬁde biooxidations. Most remarkably, opposite enantioselectivities were reported for two pheno- and genotypically similar strains of Pseudomonas frederickberensis (DSM 13022 and strain 33) [197]. Owing to their promiscuous reactivity, BVMOs were also successfully employed in sulfoxidations [198,199]. In analogy to their enantiodivergent behavior in Baeyer– Villiger oxidations, ﬂavin-dependent MOs CHMOAcineto and CPMOComa also lead to antipodal sulfoxides in whole-cell-mediated transformations [200]. Recent genome mining provided access to recombinant hydroxyacetone monooxygenase (HAPMO), which displays a very broad range of substrate acceptance for sulfoxidation reactions, in contrast to its narrow speciﬁcity upon Baeyer–Villiger oxidations [201]. Also, the thermostable PAMO from the moderately thermophilic organism Thermobiﬁda fusca represents a versatile biocatalyst for biooxidations [202], as the isolated enzyme obtained from a recombinant expression system [130] is most robust and can even be stored on the shelf. This enzyme is also capable of performing kinetic resolutions of racemic sulfoxides, as only one enantiomeric compound is converted to the corresponding sulfone with an appreciable reaction rate. Recent studies on isolated BVMOs using Rh-complexes as NADPH substitutes for facile cofactor recycling suggested a pivotal role of the native cofactor to generate the proper environment within chiral induction in sulfoxidation reactions. While biooxidation was still observed in the presence of the metal complex, stereoselectivity of the enzyme was lost almost completely [202]. Sulfoxidations are not restricted to MOs but can also be carried out by dioxygenases. For example, Pseudomonas mutant strain UV4 producing a toluene dioxygenase (TDO) and Pseudomonas NCIMB 8859 expressing a naphthalene dioxygenase (NDO) were used to oxidize aryl sulﬁdes to antipodal chiral sulfoxides [203]. In particular, the availability of such bacterial biocatalysts in the form of recombinant expression systems [136] in combination with simpliﬁed puriﬁcation protocols opened up this methodology for large-scale applications [204]. Parallel to the modiﬁcation of the catalytic performance in Baeyer–Villiger oxidations, random mutagenesis was successfully applied to improve the stereoselectivity of CHMOAcineto in cases of essentially racemic sulfoxide formation. In addition, enantiodivergent clones with >98% ee for both antipodal products were identiﬁed (Table 9.5) [205]. However, improvement in stereoselectivity of mutant enzymes was often accompanied by increased formation of sulfone. This effect can also be utilized to resolve racemic sulfoxides. Biocatalytic access to both antipodal sulfoxides was exploited in total syntheses of bioactive compounds, which is outlined in some representative examples. Biooxidation of functionalized dialkyl sulﬁdes was utilized in the direct synthesis of both enantiomers of sulforaphane and some analogs in low to good yields and stereoselectivities (Scheme 9.27) [206]. This natural product originates from broccoli and represents a potent inducer of detoxiﬁcation enzymes in mammalian metabolism; it might be related to anticarcinogenic properties of plants from the cruciform family. All four possible stereoisomers of methionine (R ¼ Me) and ethionine sulfoxides

9.6 Heteroatom Oxidations Table 9.5 Random mutagenesis of CHMOAcineto toward enantiodivergent sulfoxidations.

Sulfoxide

O S

Mutant

Amino acid modificationsa

Wild type 1-D10-F6 1-K15-C1 1-C5-H3 1-H8-A1

— D384H F432S K229I, L248P Y132C, F246I, V361A, T415A

Yield sulfoxide

eeb

Yield sulfone

75% 75% 55% 77% 52%

14% (R) 99% (R) 99% (R) 98% (S) >99% (S)

<1% 8% 20% 6% 27%

a

Position and substitution of amino acids in one-letter coding. Absolute conﬁguration in parentheses.

b

(R ¼ Et) possessing a diverse range of physiological properties were prepared in a chemoenzymatic strategy using the fungi B. bassiana and Beauveria caledonica [207]. Both enantiomers of modaﬁnil (R00 ¼ NH2) were prepared by biocatalytic sulfoxidation (Scheme 9.27), which represents a psychostimulating agent for treatment of excessive daytime sleepiness. Biotransformation of a carboxylate (R0 ¼ OH) or amid (R0 ¼ NH2) substrates using fungal strains and recombinant E. coli expressing various BVMOs gave (þ)- and ()-modaﬁnil, either directly or via the oxidized precursor (R00 ¼ OH), in moderate to excellent optical purities and chemical yields. A remarkable enzymatic cascade reaction was observed when applying Amycolapsis orientalis or Bacillus subtilis var. niger for the biotransformation: starting from the carboxylate substrate (R0 ¼ OH) sulfoxidation was accompanied by concomitant amidation to directly access (þ)- and ()-modaﬁnil, though in moderate stereoselectivities [208]. In addition to simple sulﬁdes, enzymes are also capable of interconverting other oxidation stages of sulfur. A novel catalytic activity was reported for CHMOAcineto by

NCS

S

O

Fungal strains

NCS

S

Sulforaphane

R

COOH

S

R = Me, Et

NPhth

O

Beauveria sp. R

COOH

S

NPhth

Methionine (R=Me) Ethionine (R=Et) O Ph

S Ph

Fungal strains R' Recombinant-E.coli

O Ph

S

O R''

Ph Modafinil (R''=NH2)

Scheme 9.27 Synthetic exploitation of enantiocomplementary biocatalytic sulfoxidations.

j255

j 9 Biooxidations in Chiral Synthesis

256

the oxidation of sulﬁtes. Racemic substrates were converted to sulfates in a kinetic resolution process in moderate to good selectivities for both chiral sulﬁtes and sulfates [209]. 9.6.2 Oxidation of Dithio-Compounds

Within the biooxidation of disulﬁdes, chiral thiosulﬁnates become available. TertButyl tert-butanethiosulﬁnate represents a particularly valuable chiral auxiliary for the preparation of several chiral sulfoxides and sulﬁnimines, which can be subsequently transformed into branched amine compounds, b-aminoacids, and chiral aziridines. This product is accessible readily by CHMOAcineto mediated biooxidation of tert-butyl disulﬁde to give the R-sulﬁnate in 97% ee using isolated enzyme (Scheme 9.28) [210]. Cyclic dithioketals and acetals represent another important class of sulfur containing chiral auxiliaries, which are available in chiral form by biooxidation. Biotransformations were performed on a preparative scale using whole-cells (wild type and recombinant) and isolated enzyme. Again, enantiocomplementary oxidation of unsubstituted dithianes (linear and cyclic, R ¼ H) was observed when using CHMOAcineto and CPMOComa (Scheme 9.28) [211,212]. Oxygenation of functionalized substrates (R ¼ substituted alkyl) with CHMOAcineto gave preferably trans products in moderate to good yields and enantioselectivities [213,214]. 9.6.3 Oxidation of other Heteroatoms

A successful case study for asymmetric nitrogen oxidation was reported for a series of (hetero)aromatic tertiary amines. High diastereoselectivity was observed for the enzyme-mediated oxidation of S-()-nicotine by isolated CHMOAcineto to give the corresponding cis-N-oxide [215]. The stereoselectivity of this biooxidation was complementary to the product obtained by ﬂavin MO (FMO) from human liver (trans-selective [216]) as well as unspeciﬁc oxidations by FMOs from porcine and guinea pig liver. BVMOs were also reported to facilitate mild and chemoselective conversion of boronic acids to borates, which usually hydrolyze upon biotransformation conditions using isolates protein [217]. Additionally, selenium oxidation has been described in analogy to sulfoxidations [218]. O S

CHMOAcineto

S

S

S

BVMOs

S

S R

S

S R

Scheme 9.28 Biooxidation of dithio-compounds.

O

9.7 Aryl Dihydroxylations

9.7 Aryl Dihydroxylations

Aryl dioxygenases (DOs), also referred to as Rieske-type nonheme iron oxygenases [219], are capable of converting benzene rings in a dearomatization reaction with concomitant stereoselective introduction of two hydroxyl groups [220]. The overall process involves transfer of two electrons to molecular oxygen as mild oxidant, generating a highly activated oxidizing agent coordinated to a mononuclear iron center. The dioxygenase system consists of three components forming an electron transfer cascade; initially, a ﬂavoprotein reductase is reduced at the expense of NAD(P) H. A subsequent ferredoxin domain transports electrons to the actual oxygenase, which is frequently encoded by two structural genes and contains an iron–sulfur cluster and a mononuclear Fe(II) center. The biooxygenation process commences via an intermediate dioxetane, which is cleaved enzymatically to the chiral cis-diol product. Metabolic pathways containing dioxygenases in wild-type strains are usually related to detoxiﬁcation processes upon conversion of aromatic xenobiotics to phenols and catechols, which are more readily excreted. Within such pathways, the intermediate chiral cis-diol is rearomatized by a dihydrodiol–dehydrogenase. While this mild route to catechols is also exploited synthetically [221], the chirality is lost. In the context of asymmetric synthesis, such further biotransformations have to be prevented, which was initially realized by using mutant strains deﬁcient in enzymes responsible for the rearomatization. Today, several dioxygenases with complementary substrate proﬁles are available, as outlined in Table 9.6. Considering the delicate architecture of these enzyme complexes, recombinant whole-cell-mediated biotransformations are the only option for such conversions. E. coli is preferably used as host and fermentation protocols have been optimized [222,223]. The synthetic elaboration of cis-dioxygenated dienes has ﬂourished in the past two decadesandseveraltotalsyntheseshavebeenaccomplishedutilizingthesechiralbuilding blocks. Both substrate acceptance and applications in asymmetric synthesis have been comprehensively reviewed, also providing a historical perspective [220,223,224]. 9.7.1 Dioxygenation in ortho- and meta-Position

The most abundantly utilized type of aryl dioxygenases (2,3-dioxygenases) incorporates the additional oxygen functionalities in ortho- and meta-position to an existing substituent at the aromatic core, exclusively in cis-conﬁguration, usually of very high optical purity. There are three different prototype enzymes distinguished by the nature of the preferably converted aromatic ring system (Scheme 9.29): toluene (TDO, from P. putida F39/D), naphthalene (NDO, from P. putida 119), and biphenyl dioxygenases (BPDO, from Sphingomonas yanoikuyae B8/36). Biooxidation products originating from simple homoaromatic precursors are usually formed in high regio- and stereoselectivity. Lipophilic substituents are most readily converted and the incorporation of a halogen (R ¼ Cl, Br, I) into biooxidation products of TDO is also exploited to give a more pronounced difference in reactivity

j257

j 9 Biooxidations in Chiral Synthesis

258

Table 9.6 Whole-cell strains for aryl dihydroxylations.

Bioccatalyst Dioxygenase/origin

Mutant

Recombinant

Representative substrate profile

Toluene dioxygenase (TDO) Pseudomonas putida F39/D Naphthalene dioxygenase (NDO) Pseudomonas putida 119 Naphthalene dioxygenase (NDO) Pseudomonas ﬂuorescence Biphenyl dioxygenase (BPDO) Sphingomonas yanoikuyae B8/36 Chlorobenzene dioxygenase (CDO) Pseudomonas P51 Benzoate dioxygenase (BZDO) Ralstonia eutropha (former: Alcaligenes eutrophus) Naphthalene carboxylate dioxygenase Pseudomonas putida U103

1970 [225]

1989 [226]

[224]

1975 [227]

1993 [228]

[224]

1997 [229]

[230]

1973 [231]

1989 [232]

[224]

1996 [233]

2004 [234]

[234,235]

1971 [236]

[224]

1976 [237]

[224]

of the two double bonds. The related chlorobenzene dioxygenase (CDO) is particularly suitable to convert pseudohalides (R ¼ CN, [235]) or vinylogously extended functional groups (R ¼ CH ¼ CHCN [234]). NDO is more suitable for medium-sized aromatic systems; BPDO is a particularly good biocatalyst for the dioxygenation of large aromatic molecules incorporating bay and fjord structural motifs [238]. If the substituent at the aromatic core is susceptible to oxidation like sulfur (R ¼ S-alkyl or R

R TDO

OH OH OH OH

NDO

BPDO OH OH Scheme 9.29 Representative biooxidations by prototype dioxygenases.

9.7 Aryl Dihydroxylations

OH

O

HO BPDO

Protection

O

O OH

j259

O

BPDO HO 90% yield > 98% ee

Scheme 9.30 Chemoenzymatic double dihydroxylation sequence.

S-aryl), concomitant sulfoxidation can be observed and the diastereoselectivity relative to the cis-diol depends on the nature of the sulfur substituents [239]. Usually, the polarity of compounds increases by dihydroxylation to such an extent that re-entry into the active site of dioxygenases is not possible and only single microbial biooxygenation is observed. However, when diols are masked in the form of lipophilic acetonides, a second biooxygenation reaction is possible; the diastereoselectivity of both cis-diol systems was reported to be trans (Scheme 9.30) [240]. The site of dihydroxylation in heterocycles depends on the nature of the heteroaromatic system (Scheme 9.31): usually, electron-rich heterocycles like thiophene are readily biooxidized but give conformationally labile products, which may undergo concomitant sulfoxidation [241]. Electron deﬁcient systems are not accepted; only pyridone derivatives give corresponding cis-diols [242]. Such a differentiated behavior is also observed for benzo-fused compounds: biotransformation of benzo[b] thiophene gives dihydroxylation at the heterocyclic core as major product, while quinoline and other electron-poor systems are oxidized at the homoaromatic core, predominantly [243,244]. Dioxygenases that are exploited at present do not provide general access to antipodal cis-diols and a combined chemoenzymatic strategy is required. While oxidation of bromobenzene (and other mono- and dihalogenated precursors) with TDO expressing cells gives the corresponding ()-dihydrodiol in excellent OH TDO

OH

S

S

NDO N

O

HO HO

N

O

NDO R

N

R

OH

N OH

Scheme 9.31 Biooxidation of selected heterocyclic systems.

j 9 Biooxidations in Chiral Synthesis

260

stereoselectivity, bromo-iodobenzene is converted in very poor optical purity by whole-cells expressing TDO. However, chemical reduction of the iodine in the biooxidation product leads to an almost racemic intermediate, which can then be converted with wild-type cells of Pseudomonas NCIMB 8859 with an intact metabolic pathway for aryl degradation (Scheme 9.32). The (þ)-intermediate is accepted as natural substrate and completely metabolized while ()-dihydrodiol is recovered in high optical purity [245]. Proper adaptation of the synthetic elaboration with respect to the order of reaction steps and utilizing the different reactivities of double bonds depending on the nature of the substituent R (Scheme 9.29) has also been outlined as efﬁcient strategy to access enantiocomplementary target compounds [246]. This concept of exploiting the pro-enantiotopic relationship between biooxidation product and target compounds was also applied in the ﬁrst utilization of a single cisdihydroxydiene for the total synthesis of both enantiomers of pinitol [247,248]. Within the diastereomeric switch sequences, the corresponding trans-diols become accessible either using a Mitsunobu inversion or a reversible Diels–Alder cyclization as key reaction step [249,250]. This synthetic strategy is complementary to an approach involving metabolic engineering of E. coli via the chorismate/ isochorismate pathway [251]. Application of chiral cis-dihydroxydiene building blocks in the synthesis of natural products is legion. Recurring targets are polyoxygenated cycloalkanes such as inositols [252], modiﬁed sugars [253], as well as diverse alkaloids, in particular, the morphine scaffold [254]. Some of the general synthetic strategies and key transformations applied in asymmetric total synthesis are outlined with recent examples (Scheme 9.33); in the synthesis of carba-sugars, sequential biocatalytic and chemical oxygenation steps are followed by a metal-assisted functionalization to introduce the C-6 group at the ﬁnal galactose-analog [255].

I

Br

Br OH

TDO Br

(+)-diol 99% ee

OH

TDO I OH

OH

H2/Pd(C)

OH Br 22% ee

P. putida NCIMB 8859

OH

OH Br (−)-diol

OH Br

22% ee Scheme 9.32 Chemoenzymatic strategy to enantiocomplementary cis-diols.

(−)-diol 99% ee

9.7 Aryl Dihydroxylations

j261

Cycloaddition strategies allow for highly regioselective 1,4-functionalization of the diene system, in particular in combination with hetero-dienophiles. 1,4-Hydroxyamines obtained via this concept then allow access to aziridines, which can undergo CC bond formation within the ring-opening of the strained systems. The resulting nitrogen functionality was ultimately utilized in a Bischler–Napieralski-type

I

X

I OR

TDO

Osmylation

OH

OR

HO

OR

CH2OH OR

OH

HO

OH

R=H R=CMe2

Metalation

OH

X=I

Carba-α-L-galactopyranose

X=COOMe OH OR

Hetero DielsAlder cyclization

OR PG N

OR

OR OR

O

OR N H PG

OH HO

OH

Bischler-Napieralski cyclization

OH

RO

Cross-coupling aziridin-opening

OR

aryl

NH

RO

OR

OR OR

N

H N COOM e

COOM e

O Pancratistatin Cl

Cl

Diastereomeric switch

OH

OPG

OH

OPG

Alkene cleavage

M eOOC OH C

OPG OPG

O O OH OH

O

OPG

Ring closing metathesis

O O

OPG

Cladospolide C Scheme 9.33 Synthetic strategy toward carba-sugars, Amaryllidaceae alkaloids, and undecenolides.

OPG OPG

j 9 Biooxidations in Chiral Synthesis

262

cyclization reaction (after eventual chemical oxygenation) to access an array of Amaryllidaceae alkaloids of the pancratistatin type [256,257]. The diene core does not necessarily need to remain intact, but can rather be cleaved to enable further functionalization. Taking advantage of the diastereomeric switch, such a strategy was implemented together with a ring closing metathesis step in the total synthesis of undecenolides like cladospolide [258,259]. The catalytic repertoire of dioxygenases was expanded by directed evolution utilizing a gene shufﬂing methodology with the intention of improving dihydroxylation performance for p-xylene. Multigene DNA shufﬂing of TDO and tetrachlorobenzene dioxygenase generated chimera biocatalysts capable of producing up to 20 g l1 of the corresponding diol. This intermediate was then converted chemically to ﬂavor compound strawberry furanone (Scheme 9.34) [260]. OH

O O Evolved dioxygenase

OH

OH

Ozonolysis +

OH

HO

O

OH

O MeO

OH

Cyclization

OH

O Strawberry furanone

Scheme 9.34 Synthesis of a flavor compound by genetically evolved dioxygenases.

9.7.2 Dioxygenation in ipso- and ortho-Position

In contrast to 2,3-dioxygenases, the related ipso/ortho oxygenation of aryl carboxylates has received considerable less attention and has hardly been utilized by the synthetic community, so far. Biooxidation of benzoic acid and b-naphthalene carboxylate provide access to corresponding 1,2-dihydroxylated dihydroaryl compounds in excellent stereoselectivity (Scheme 9.35), analogous to TDO- and NDOmediated ortho/meta oxygenations. Whole-cell-mediated biotransformations were performed with mutant strains of Ralstonia and Pseudomonas and enable access to preparative quantities in >5 g l1 titers [261,262]. Biooxidized benzoic acid is susceptible to several transformations. Chemical hydroxylation can be directed depending on the size of the hydroxyl protecting groups [262]. Such a chemoenzymatic oxygenation sequence was implemented in the total synthesis of a carba-analog (X ¼ CH2) of the drug topiramate (X ¼ O), which is used in the treatment of epileptic seizures and migraine prophylaxis (Scheme 9.36) [263]. The diene system can undergo Diels–Alder additions; however, the regioselectivity is limited and only symmetrical dienophiles or intramolecular cyclizations give good results [264,265].

9.8 Halogenation Reactions

COOH

HOOC Ralstonia sp.

OH OH

OH COOH

j263

OH

Pseudomonas sp.

COOH

Scheme 9.35 Biodihydroxylation of aryl carboxylates by mutant whole-cells.

COOH

ROOC Biooxygenation

O

O

O

ROOC Osmylation

Protection

O OH

OH

H2NSO2O

O X

X = CH2, O

O O

O

Topiramate (X = O)

Scheme 9.36 Application of ipso/ortho dihydroxylation in the synthesis of a topiramate analog.

9.8 Halogenation Reactions

Biocatalytic halogenation reactions are found on a large structural diversity of natural substrates and an increasing number of halogenated natural products have been found to date [266–268]. Nature seems to utilize enzymatic halogen incorporation to modify physiochemical properties in order to ﬁne-tune the afﬁnity of certain secondary metabolites versus biopolymers. Although the introduction of ﬂuorine and iodine is hardly observed in natural products, organochlorides and organobromides are predominantly found in halogenated natural products. It is interesting to note that chloro-compounds are abundantly encountered in terrestrial microorganisms, while bromides are preferred as biohalogenation metabolites in marine species. Enzyme-mediated halogenation is mostly achieved by oxidative strategies and four types of enzymes play a major role with different prosthetic groups in the active center [269]. Heme-dependent haloperoxidases generate HOX as reactive species from H2O2 and X, which represents an Xþ equivalent capable of undergoing electrophilic addition at electron-rich centers [270,271]. A prototype biocatalyst of this group is the chloroperoxidase from Caldariomyces fumago [272]. In many natural systems, such enzymes are responsible for the halogenation of electron-rich aromatic cores. Similarly, ﬂavin-dependent halogenases also generate an HOX equivalent as electrophilic species for reactions. The mechanism was studied using tryptophan7-halogenase and is reported to proceed via a peroxyﬂavin species with O2 acting as oxidant similar to ﬂavin monooxygenases [273]. Again, such systems are efﬁcient biocatalysts for the regioselective halogenation of (hetero)aromatic substrates.

j 9 Biooxidations in Chiral Synthesis

264

HO Nerolidol Br-, H2O2

VBPO

Br

OH

+ R

+ Br

R

Br

Br

OH

Br

OH

Scheme 9.37 Vanadium-dependent bromoperoxidase (VBPO)mediated conversion of nerolidol to a-, b-, and g-snyderols.

In vanadium-dependent haloperoxidases, the metal center is coordinated to the imidazole system of a histidine residue, which is similarly responsible for creating hypochlorite or hypobromite as electrophilic halogenating species [274]. Remarkably, a representative of this enzyme class is capable of performing stereoselective incorporation of halides, as has been reported for the conversion of nerolidol to various snyderols. The overall reaction commences through a bromonium intermediate, which cyclizes in an intramolecular process; the resulting carbocation can ultimately be trapped upon elimination to three snyderols (Scheme 9.37) [275]. The group of nonheme iron a-ketoglutarate-dependent halogenases is remarkable in that these enzymes are able to halogenate nonactivated carbon centers [276]. Such biocatalysts utilize high-valent oxoiron species similar to P450 enzymes (sharing also some structural features with such proteins [277]) and are in conjunction with large nonribosomal peptide synthetase machineries. Multiple turnovers of such a sequence can result in polyhalogenation like the stereoselective trichlorination of a leucine methyl group within the formation of barbamide. Various stages of such a repetitive process have been studied (Scheme 9.38) [278,279]. A more comprehensive evaluation of the applicability can be expected from the recent cloning of such enzymes in E. coli and P. putida [280]. Cl

Cl

CCl3

SH

+ H2N

COOH

BarB2 BarA

H2N

COS

BarB1 H2N

COS

BarA

Scheme 9.38 Trichlorination of leucine with enzymes from the barbamide biosynthesis pathway (BarB1 & BarB2 represent nonheme iron ketoglutarate-dependent halogenases).

BarA

H2N

COS BarA

References

Considering the tremendous progress made in recent years to understand and utilize enzymatic halogenations in stereoselective synthesis, this area will continue to receive signiﬁcant attention and interesting new developments can be expected in the years to come.

9.9 Summary and Outlook

As a consequence of the revolution in genome deciphering in recent years, a large number of novel biooxidation catalysts have been and still are becoming available. The remarkable diversity of biotransformations enabled by this highly diverse group of enzymes together with the high promiscuity of several of such biocatalysts makes these proteins highly appealing catalytic entities for applications in asymmetric synthetic chemistry. For many types of reactions, enzyme platforms have been proposed and characterized, allowing entry to a multitude of interesting building blocks with high chemo-, regio-, and enantioselectivity, which in many cases provide access to enantiocomplementary products either directly or by subsequent chemical methods. While the utilization of these rather complex enzymes is still in part compromised by their accessibility, limited stability, and cofactor requirements, several interesting strategies have been outlined to enable application of such catalytic systems in synthetic laboratories or even on production scale. Certainly, the ﬁeld is lacking in the simple protocols and robust catalysts available in some other areas of biocatalysis; however, this is largely compensated by the exciting types of transformations enabled within biooxidations. Moreover, the arsenal of interesting reactions mediated by oxidative enzymes is continuously expanding and new biotransformations are expected to be discovered within this rapidly emerging section of enzyme catalysis.

References 1 Buhler, B. and Schmid, A. (2004) Journal of Biotechnology, 113, 183–210. 2 Zhao, H. and van der Donk, W.A. (2003) Current Opinion in Biotechnology, 14, 583–589. 3 Hollmann, F., Hofstetter, K. and Schmid, A. (2006) Trends in Biotechnology, 24, 163–171. 4 Hollmann, F. and Schmid, A. (2004) Biocatalysis and Biotransformation, 22, 63–88. 5 Hollmann, F., Lin, P.-C., Witholt, B. and Schmid, A. (2003) Journal of the American Chemical Society, 125, 8209–8217.

6 D. ArrigoDArrigo, P., Pedrocchi-Fantoni, G. and Servi, S. (2000) Stereoselective Biocatalysis, 1, 365–396. 7 Stewart, J.D. (2000) Current Opinion in Biotechnology, 11, 363–368. 8 Lehmann, M. and Wyss, M. (2001) Current Opinion in Biotechnology, 12, 371–375. 9 Drauz, K. and Waldmann, H. (2002) Enzyme Catalysis in Organic Synthesis, Wiley-VCH Verlag GmbH, Weinheim. 10 Bommarius, A.S. and Riebel, B.R. (2004) Biocatalysis, Wiley-VCH Verlag GmbH, Weinheim.

j265

j 9 Biooxidations in Chiral Synthesis

266

11 Patel, R.N. (2007) Biocatalysis in the Pharmaceutical and Biotechnology Industries, CRC Press, Boca Raton. 12 Li, Z., van Beilen, J.B., Duetz, W.A., Schmid, A., de Raadt, A., Griengl, H. and Witholt, B. (2002) Current Opinion in Chemical Biology, 6, 136–144. 13 Burton, S.G. (2003) Trends in Biotechnology, 21, 543–549. 14 van Beilen, J.B., Duetz, W.A., Schmid, A. and Witholt, B. (2003) Trends in Biotechnology, 21, 170–177. 15 B€ uhler, B. and Schmid, A. (2004) Journal of Biotechnology, 113, 183–210. 16 Azerad, R. and Buisson, D. (2000) Current Opinion in Biotechnology, 11, 565–571. 17 Turner, N.J. (2004) Current Opinion in Biotechnology, 8, 114–119. 18 Kroutil, W., Mang, H., Edegger, K. and Faber, K. (2004) Current Opinion in Chemical Biology, 8, 120–126. 19 Kroutil, W., Mang, H., Edegger, K. and Faber, K. (2004) Advanced Synthesis and Catalysis, 346, 125–142. 20 Hummel, W. (1997) Advances in Biochemical Engineering/Biotechnology, 58, 147–179. 21 Nakamura, K., Miyai, T., Nakajima, N. and Ohno, A. (1990) Tetrahedron Letters, 31, 1159–1160. 22 Lee, L.G. and Whitesides, G.M. (1986) Journal of Organic Chemistry, 51, 25–36. 23 Alexandre, F.-R., Pantaleone, D.P., Taylor, P.P., Fotheringham, I.G., Ager, D.J. and Turner, N.J. (2002) Tetrahedron Lett., 43, 707–710. 24 Beard, T.M. and Turner, N.J. (2002) Chem. Commun., 246–247. 25 Carr, R., Alexeeva, M., Enright, A., Eve, T.S.C., Dawson, M.J. and Turner, N.J. (2003) Angew. Chem. Int. Ed., 42, 4807–4810. 26 Secondo, F., Carrea, G., De Amici, M., Di Ventimiglia, S.J. and Dordick, J.S. (2003) Biotechnology and Bioengineering, 81, 391–396. 27 Riva, S., Bovara, R., Pasta, P. and Carrea, G. (1986) Journal of Organic Chemistry, 51, 2902–2906.

28 Riva, S., Bovara, R., Zetta, L., Pasta, P., Ottolina, G. and Carrea, G. (1988) Journal of Organic Chemistry, 53, 88–92. 29 Anelli, P.L., Bocchetta, M., Morosini, P., Palano, D., Carrea, G., Falcone, L., Pasta, P. and Sartore, D. (2002) Biocatalysis and Biotransformation, 20, 29–34. 30 Drueckhammer, D.G., Hennen, W.J., Pederson, R.L., Barbas, C.F., Gautheron, C.M., Krach, T. and Wong, C.-H. (1991) Synthesis, 7, 499–525. 31 Reddy, S.Y. and Bruice, T.C. (2004) Protein Science, 13, 1965–1978. 32 Giffhorn, F. (2000) Applied Microbiology and Biotechnology, 54, 727–740. 33 Leitner, C., Neuhauser, W., Volc, J., Kulbe, K.D., Nidetzky, B. and Haltrich, D. (1998) Biocatalysis and Biotransformation, 16, 365–382. 34 Haltrich, D., Leitner, C., Nidetzky, B. and Kulbe, K.D. (2000) Progress in Biotechnology, 17, 137–149. 35 Whittaker, J.W. and Whittaker, M.M. (2000) Protein Expression and Puriﬁcation, 20, 105–111. 36 Whittaker, J.W. (2003) Chemical Reviews, 103, 2347–2363. 37 Manzoni, M., Rollini, M. and Bergomi, S. (2001) Process Biochemistry, 36, 971–977. 38 Mayer, A.M. and Staples, R.C. (2002) Phytochemistry, 60, 551–565. 39 Monti, D., Candido, A., Cruz Silva, M.M., Kren, V., Riva, S. and Danieli, B. (2005) Advanced Synthesis and Catalysis, 347, 1168–1174. 40 Jakovac, I.J., Goodbrand, H.B., Lok, K.P. and Jones, J.B. (1982) Journal of the American Chemical Society, 104, 4659–4665. 41 Lok, K.P., Jakovac, I.J. and Jones, J.B. (1985) Journal of the American Chemical Society, 107, 2521–2526. 42 Gourdel-Martin, M.-E., Comoy, C., and Huet, F. (1999) Tetrahedron: Asymmetry, 10, 403–404. 43 Edegger, K., Stampfer, W., Seisser, B., Faber, K., Mayer, S.F., Oehrlein, R., Hafner, A. and Kroutil, W. (2006)

References

44

45

46

47

48

49 50 51

52

53

54

55 56

57

58

European Journal of Organic Chemistry, 8, 1904–1909. Kosjek, B., Stampfer, W., van Deursen, R., Faber, K. and Kroutil, W. (2003) Tetrahedron, 59, 9517–9521. Tecelao, C.S.R., van Keulen, F. and Fonseca, M.M.R. (2001) Journal of Molecular Catalysis. B, Enzymatic, 11, 719–724. Molinari, F., Gandolﬁ, R., Aragozzini, F., Leon, R. and Prazeres, D.M.F. (1999) Enzyme and Microbial Technology, 25, 729–735. Carnell, A.J. (1999) Advances in Biochemical Engineering/Biotechnology, 63, 57–72. Gruber, C.C., Lavandera, I., Faber, K. and Kroutil, W. (2006) Advanced Synthesis and Catalysis, 348, 1789–1805. Pamies, O. and B€ackvall, J.-E. (2003) Chemical Reviews, 103, 3247–3261. Pamies, O. and B€ackvall, J.-E. (2004) Trends in Biotechnology, 22, 130–135. Hummel, W. and Riebel, B. (1996) Annals of the New York Academy of Sciences, 799, 713–716. Adam, W., Lazarus, M., Boss, B., Saha-M€ oller, C.R., Humpf, H.U. and Schreier, P. (1997) Journal of Organic Chemistry, 62, 7841–7843. Adam, W., Lazarus, M., Saha-M€oller, C.R. and Schreier, P. (1998) Tetrahedron: Asymmetry, 9, 351–355. Cardus, G.J., Carnell, A.J., Trauthwein, H. and Riermeir, T. (2004) Tetrahedron: Asymmetry, 15, 239–243. Allan, G.R. and Carnell, A.J. (2001) Journal of Organic Chemistry, 66, 6495–6497. Comasseto, J.V., Omori, A.T., Andrade, L.H. and Porto, A.L.M. (2003) Tetrahedron: Asymmetry, 14, 711–715. Demir, A.S., Hamamci, H., Sesenoglu, O., Nselihanoglu, R., Asikoglu, B. and Capanoglu, D. (2002) Tetrahedron Letters, 43, 6447–6449. Andrade, L.H., Utsunomiya, R.S., Omori, A.T., Porto, A.L.M. and Comasseto, J.V. (2006) Journal of Molecular Catalysis. B, Enzymatic, 38, 84–90.

59 Nie, Y., Xu, Y. and Mu, X.Q. (2004) Organic Process Research and Development, 8, 246–251. 60 Fotheringham, I., Archer, O., Carr, R., Speight, R. and Turner, N.J. (2006) Biochemical Society Transactions, 34, 287–290. 61 Alexandre, F.-R., Pantaleone, D.P., Taylor, P.P., Fotheringham, I.G., Ager, D.J. and Turner, N.J. (2002) Tetrahedron Letters, 43, 707–710. 62 Beard, T.M. and Turner, N.J. (2002) Chemical Communications, 246–247. 63 Enright, A., Alexandre, F.-R., Roff, G., Fotheringham, I.G., Dawson, M.J. and Turner, N.J. (2003) Chemical Communications, 2636–2637. 64 Roff, G.J., Lloyd, R.C. and Turner, N.J. (2004) Journal of the American Chemical Society, 126, 4098–4099. 65 Carr, R., Alexeeva, M., Enright, A., Eve, T.S.C., Dawson, M.J. and Turner, N.J. (2003) Angewandte Chemie International Edition, 42, 4807–4810. 66 Dunsmore, C.J., Carr, R., Fleming, T. and Turner, N.J. (2006) Journal of the American Chemical Society, 128, 2224–2225. 67 Azerad, R. (2006) Asymmetric Oxidation Reactions, Oxford University Press, pp. 181–200. 68 Holland, H.L. and Weber, H.K. (2000) Current Opinion in Biotechnology, 11, 547–553. 69 Fernandes, P., Cruz, Y., Angelova, B., Pinheiro, H.M. and Cabral, J.M.S. (2003) Enzyme and Microbial Technology, 32, 688–705. 70 Grogan, G., Roberts, G.A., Parsons, S., Turner, N.J. and Flitsch, S.L. (2002) Applied Microbiology and Biotechnology, 59, 449–454. 71 Celik, A., Flitsch, S.L. and Turner, N.J. (2005) Organic and Biomolecular Chemistry, 3, 2930–2934. 72 Duetz, W.A., Fjaellman, A.H.M., Ren, S., Jourdat, C. and Witholt, B. (2001) Applied and Environmental Microbiology, 67, 2829–2832.

j267

j 9 Biooxidations in Chiral Synthesis

268

73 Hamada, H., Tanaka, T., Furuya, T., Rakahata, H. and Remoto, H. (2001) Tetrahedron Letters, 42, 909–911. 74 Adam, W., Lukacs, Z., Harmsen, D., Saha-M€oller, C.R. and Schreier, P. (2000) Journal of Organic Chemistry, 65, 878–882. 75 Adam, W., Lukacs, Z., Kahle, C., SahaM€oller, C.R. and Schreier, P. (2001) Journal of Molecular Catalysis. B, Enzymatic, 11, 377–385. 76 Li, Z., Feiten, H.-J., Chang, D., Duetz, W.A., van Beilen, J.B. and Witholt, B. (2001) Journal of Organic Chemistry, 66, 8424–8430. 77 Chang, D., Feiten, H.-J., Engesser, K.-H., van Beilen, J.B., Witholt, B. and Li, Z. (2002) Organic Letters, 4, 1859–1862. 78 de Raadt, A. and Griengl, H. (2002) Current Opinion in Biotechnology, 13, 537–542. 79 Braunegg, G., de Raadt, A., Feichtenhofer, S., Griengl, H., Kopper, I., Lehmann, A., and Weber, H.-J. (1999) Angewandte Chemie International Edition, 38, 2763–2766. 80 M€ unzer, D.F., Griengl, H., Moumtzi, A., Saf, R., Terzani, T. and de Raadt, A. (2005) European Journal of Organic Chemistry, 5, 793–796. 81 de Raadt, A., Griengl, H. and Weber, H. (2001) Chemistry- A European Journal, 7, 27–31. 82 Olivo, H.F. and Hemenway, M.S. (1999) Journal of Organic Chemistry, 64, 8968–8969. 83 Olivo, H.F. and Hemenway, M.S. (1999) Journal of Organic Chemistry, 64, 6312–6318. 84 Urlacher, V.B., Lutz-Wahl, S. and Schmid, R.D. (2004) Applied Microbiology and Biotechnology, 64, 317–325. 85 Urlacher, V.B. and Eiben, S. (2006) Trends in Biotechnology, 24, 324–330. 86 Urlacher, V.B. and Schmid, R.D. (2006) Current Opinion in Chemical Biology, 10, 156–161. 87 Maurer, S.C., K€ uhnel, K., Kaysser, L.A., Eiben, S., Schmid, R.D. and Urlacher, V.B. (2005) Advanced Synthesis and Catalysis, 347, 1090–1098.

88 Poulos, T.L., Finzel, B.C. and Howard, A.J. (1987) Journal of Molecular Biology, 195, 687–700. 89 Bell, S.G., Stevenson, J.-A., Boyd, H.D., Campbell, S., Riddle, A.D., Orton, E.L. and Wong, L.-L. (2002) Chemical Communications, 490–491. 90 French, K.J., Rock, D.A., Rock, D.A., Manchester, J.I., Goldstein, B.M. and Jones, J.P. (2002) Archives of Biochemistry and Biophysics, 398, 188–197. 91 Xu, F., Bell, S.G., Lednik, J., Insley, A., Rao, Z. and Wong, L.-L. (2005) Angewandte Chemie International Edition, 44, 4029–4032. 92 Speight, R.E., Hancock, R.D., Winkel, C., Bevinakatti, H.S., Sarkar, M., Flitsch, S.L. and Turner, N.J. (2004) Tetrahedron: Asymmetry, 15, 2829–2831. 93 Farinas, E.T., Schwaneberg, U., Glieder, A. and Arnold, F.H. (2001) Advanced Synthesis and Catalysis, 343, 601–605. 94 Glieder, A., Farinas, E.T. and Arnold, F.H. (2002) Nature Biotechnology, 20, 1135–1139. 95 Peters, M.W., Meinhold, P., Glieder, A. and Arnold, F.H. (2003) Journal of the American Chemical Society, 125, 13442–13450. 96 Craft, D.L., Madduri, K.M., Eshoo, M. and Wilson, C.R. (2003) Applied and Environmental Microbiology, 69, 5983–5991. 97 M€ unzer, D.F., Meinhold, P., Peters, M.W., Feichtenhofer, S., Griengl, H., Arnold, F.H., Glieder, A. and de Raadt, A. (2005) Chemical Communications, 2597–2599. 98 Blasco, F., Kauffmann, I. and Schmid, R.D. (2004) Applied Microbiology and Biotechnology, 64, 671–674. 99 Bell, S.G., Chen, X., Sowden, R.J., Xu, F., Williams, J.N., Wong, L.-L. and Rao, Z. (2003) Journal of the American Chemical Society, 125, 705–714. 100 Urlacher, V.B., Makhsumkhanov, A. and Schmid, R.D. (2006) Applied Microbiology and Biotechnology, 70, 53–59. 101 Sowden, R.J., Yasmin, S., Rees, N.H., Bell, S.G. and Wong, L.-L. (2005) Organic and Biomolecular Chemistry, 3, 57–64.

References 102 Croteau, R.B. and W€ ust, M. (2002) Biochemistry, 41, 1820–1827. 103 Novak, M.J. (1999) Bioorganic and Medicinal Chemistry Letters, 9, 31–34. 104 Yadav, J.S., Nanda, S. and Rao, A.B. (2001) Tetrahedron: Asymmetry, 12, 2129–2135. 105 Chang, D., Zhang, J., Witholt, B. and Li, Z. (2004) Biocatalysis and Biotransformation, 22, 113–130. 106 Sharpless, K.B. (2002) Angewandte Chemie International Edition, 41, 2024–2032. 107 Park, J.-B., B€ uhler, B., Habicher, T., Hauer, B., Panke, S., Witholt, B. and Schmid, A. (2006) Biotechnology and Bioengineering, 95, 501–512. 108 Hollmann, F., Lin, P.-C., Witholt, B. and Schmid, A. (2003) J. Am. Chem. Soc., 125, 8209–8217. 109 Steinig, G.H., Livingston, A.G. and Stuckey, D.C. (2000) Biotechnology and Bioengineering, 70, 553–563. 110 Panke, S., Held, M., Wubbolts, M.G., Witholt, B. and Schmid, A. (2002) Biotechnology and Bioengineering, 80, 33–41. 111 Guengerich, F.P. (2003) Archives of Biochemistry and Biophysics, 409, 59–71. 112 Dembitsky, V.M. (2003) Tetrahedron, 59, 4701–4720. 113 Wubbolts, M.G., Panke, S., van Beilen, J.B. and Witholt, B. (1996) Chimia, 50, 436–437. 114 Schmid, A., Hofstetter, K., Feiten, H.-J., Hollmann, H. and Witholt, B. (2001) Advanced Synthesis and Catalysis, 343, 732–737. 115 Hu, S., Gupta, P., Prasad, A.K., Gross, R.A. and Parmar, V.S. (2002) Tetrahedron Letters, 43, 6763–6766. 116 Staijen, I.E., van Beilen, J.B. and Witholt, B. (2000) European Journal of Biochemistry, 267, 1957–1965. 117 Hu, S. and Hager, L.P. (1999) Tetrahedron Letters, 40, 1641–1644. 118 Lakner, F.J. and Hager, L.P. (1996) Journal of Organic Chemistry, 61, 3923–3925. 119 Chang, D., Heringa, M.F., Witholt, B. and Li, Z. (2003) Journal of Organic Chemistry, 68, 8599–8606.

120 Hilker, I., Bothe, D., Pr€ uss, J. and Warnecke, H.-J. (2001) Chemical Engineering Science, 56, 427–432. 121 Zhao, L.-Q., Sun, Z.-H., Zheng, P. and Zhu, L.-L. (2005) Biotechnology Letters, 27, 1505–1509. 122 Mang, H., Gross, J., Lara, M., Goessler, C., Schoemaker, H.E., Guebitz, G.M. and Kroutil, W. (2006) Angewandte Chemie International Edition, 45, 5201–5203. 123 Mihovilovic, M.D., Rudroff, F. and Gr€otzl, B. (2004) Current Organic Chemistry, 8, 1057–1069. 124 van Berkel, W.J.H., Kamerbeek, N.M. and Fraaije, M.W. (2006) Journal of Biotechnology, 124, 670–689. 125 Mihovilovic, M.D. (2006) Current Organic Chemistry, 10, 1265–1287. 126 Kamerbeek, N.M., Janssen, D.B., van Berkel, W.J.H. and Fraaije, M.W. (2003) Advanced Synthesis and Catalysis, 345, 667–678. 127 Mihovilovic, M.D., M€ uller, B. and Stanetty, P. (2002) European Journal of Organic Chemistry, 3711–3730. 128 Ryerson, C.C., Ballou, D.P. and Walsh, C. (1982) Biochemistry, 21, 2644–2655. 129 Sheng, D., Ballou, D.P. and Massey, V. (2001) Biochemistry, 40, 11156–11167. 130 Fraaije, M.W., Wu, J., Neuts, D.P.H.M., van Hellemond, E.W., Spelberg, J.H.L. and Janssen, D.B. (2005) Applied Microbiology and Biotechnology, 66, 393–400. 131 Malito, E., Alﬁeri, A., Fraaije, M.W. and Mattevi, A. (2004) Proceedings of the National Academy of Sciences of the United States of America, 101, 13157–13162. 132 Stewart, J.D. (1998) Current Organic Chemistry, 2, 195–216. 133 Fraaije, M.W., Kamerbeek, N.M., van Berkel, W.J.H. and Janssen, D.B. (2002) FEBS Letters, 518, 43–47. 134 van Beilen, J.B., Mourlane, F., Seeger, M.A., Kovac, J., Li, Z., Smits, T.H.M., Fritsche, U. and Witholt, B. (2003) Environmental Microbiology, 5, 174–182.

j269

j 9 Biooxidations in Chiral Synthesis

270

135 Stewart, J.D., Reed, K.W. and Kayser, M.M. (1996) Journal of the Chemical Society. Perkin Transactions 1, 8, 755–757. 136 Chen, G., Kayser, M.M., Mihovilovic, M.D., Mrstik, M.E., Martinez, C.A. and Stewart, J.D. (1999) New Journal of Chemistry, 23, 827–832. 137 Iwaki, H., Wang, S., Grosse, S., Bergeron, H., Nagahashi, A., Lertvorachon, J., Yang, J., Konishi, Y., Hasegawa, Y. and Lau, P.C.K. (2006) Applied and Environmental Microbiology, 72, 2707–2720. 138 Mihovilovic, M.D., Snajdrova, R., Winninger, A. and Rudroff, F. (2005) Synlett, 18, 2751–2754. 139 Kirschner, A. and Bornscheuer, U.T. (2006) Angewandte Chemie International Edition, 45, 7004–7006. 140 Bonsor, D., Butz, S.F., Solomons, J., Grant, S., Fairlamb, I.J.S., Fogg, M.J. and Grogan, G. (2006) Organic and Biomolecular Chemistry, 4, 1252–1260. 141 Snajdrova, R., Grogan, G. and Mihovilovic, M.D. (2006) Bioorganic and Medicinal Chemistry Letters, 16, 4813–4817. 142 Kostichka, K., Thomas, S.M., Gibson, K.J., Nagarajan, V. and Cheng, Q. (2001) Journal of Bacteriology, 183, 6478–6486. 143 Kyte, B.G., Rouviere, P., Cheng, Q. and Stewart, J.D. (2004) Journal of Organic Chemistry, 69, 12–17. 144 Donoghue, N.A., Norris, D.B. and Trudgill, P.W. (1976) European Journal of Biochemistry, 63, 175–192. 145 Chen, Y.-C.J., Peoples, O.P. and Walsh, C.T. (1988) Journal of Bacteriology, 170, 781–789. 146 Stewart, J.D. (1998) Curr. Org. Chem., 2, 195–216. 147 Brzostowicz, P., Walters, D.M., Thomas, S.M., Nagarajan, V. and Rouviere, P.E. (2003) Applied Environmental Microbiology, 69, 334–342. 148 Mihovilovic, M.D., Rudroff, F., Gr€otzl, B., Kapitan, P., Snajdrova, R., Rydz, J. and Mach, R. (2005) Angewandte Chemie International Edition, 44, 3609–3613.

149 Bramucci, M.G., Brzostowicz, P.C., Kostichka, K.N., Nagarajan, V., Rouviere, P.E. and Thomas, S.M. (2003) Patent Cooperation Treaty International Application, WO 2003020890; Chemical Abstracts, 138, 233997. 150 Brzostowicz, P.C., Gibson, K.L., Thomas, S.M., Blasko, M.S. and Rouviere, P.E. (2000) Journal of Bacteriology, 182, 4241–4248. 151 Mihovilovic, M.D., Rudroff, F., M€ uller, B. and Stanetty, P. (2003) Bioorganic & Medicinal Chemistry Letters, 13, 1479–1482. 152 Grifﬁn, M. and Trudgill, P.W. (1976) European Journal of Biochemistry, 63, 199–209. 153 Iwaki, H., Hasegawa, Y., Lau, P.C.K., Wang, S. and Kayser, M.M. (2002) Applied Environmental Microbiology, 68, 5681–5684. 154 Fraaije, M.W., Kamerbeek, N.M., Heidekamp, A.J., Fortin, R. and Janssen, D.B. (2004) J. Biol. Chem., 279, 3354–3360. 155 Walton, A.Z. and Stewart, J.D. (2002) Biotechnology Progress, 18, 262–268. 156 Hilker, I., Alphand, V., Wohlgemuth, R. and Furstoss, R. (2004) Advanced Synthesis and Catalysis, 346, 203–214. 157 Rudroff, F., Alphand, V., Furstoss, R. and Mihovilovic, M.D. (2006) Organic Process Research and Development, 10, 599–604. 158 Hilker, I., Wohlgemuth, R., Alphand, V. and Furstoss, R. (2005) Biotechnology and Bioengineering, 92, 702–710. 159 Mihovilovic, M.D., M€ uller, B., Kayser, M.M., Stewart, J.D., Fr€ohlich, J., Stanetty, P. and Spreitzer, H. (2001) Journal of Molecular Catalysis. B, Enzymatic, 11, 349–353. 160 Mihovilovic, M.D., Rudroff, F., Kandioller, W., Gr€otzl, B., Stanetty, P. and Spreitzer, H. (2003) Synlett, 1973–1976. 161 Mihovilovic, M.D., Bianchi, D.A. and Rudroff, F. (2006) Chemical Communications, 3214–3216. 162 Taschner, M.J. and Black, D.J. (1988) Journal of the American Chemical Society, 110, 6892–6893.

References 163 Ottolina, G., de Gonzalo, G., Carrea, G. and Danieli, B. (2005) Advanced Synthesis and Catalysis, 347, 1035–1040. 164 Braun, I., Rudroff, F., Mihovilovic, M.D. and Bach, T. (2006) Angewandte Chemie International Edition, 45, 5541–5543. 165 Alphand, V., Mazzini, C., Lebreton, J. and Furstoss, R. (1998) Journal of Molecular Catalysis. B, Enzymatic, 5, 219–221. 166 Reetz, M.T., Brunner, B., Schneider, T., Schulz, F., Clouthier, C.M. and Kayser, M.M. (2004) Angewandte Chemie International Edition, 43, 4075–4078. 167 Mihovilovic, M.D., Rudroff, F., Winninger, A., Schneider, T., Schulz, F. and Reetz, M.T. (2006) Organic Letters, 8, 1221–1224. 168 Clouthier, C.M., Kayser, M.M. and Reetz, M.T. (2006) Journal of Organic Chemistry, 71, 8431–8437. 169 Alphand, V., Furstoss, R., PedragosaMoreau, S., Roberts, S.M. and Willetts, A.J. (1996) Journal of the Chemical Society, Perkin Transactions 1, 1867–1872. 170 Stewart, J.D., Kieth, W.R., Zhu, J., Chen, G. and Kayser, M.M. (1996) Journal of Organic Chemistry, 61, 7652–7653. 171 Kayser, M.M., Chen, G. and Stewart, J. (1998) Journal of Organic Chemistry, 63, 7103–7106. 172 Stewart, J.D., Reed, K.W., Martinez, C.A., Zhu, J., Chen, G. and Kayser, M.M. (1998) Journal of the American Chemical Society, 120, 3541–3548. 173 Wang, S., Chen, G., Kayser, M.M., Iwaki, H., Lau, C.K. and Hasegawa, Y. (2002) Canadian Journal of Chemistry, 80, 613–621. 174 Bes, M.T., Villa, R., Roberts, S.M., Wan, P.W.H. and Willetts, A.J. (1996) Journal of Molecular Catalysis. B, Enzymatic, 1, 127–134. 175 Adger, B.M., McCague, R. and Roberts, S.M. (1996) International Patent Application WO 96/38437, Chemical Abstracts, 126, 89208.

176 Alphand, V., Archelas, A. and Furstoss, R. (1990) Journal of Organic Chemistry, 55, 347–350. 177 Berezina, N., Alphand, V. and Furstoss, R. (2002) Tetrahedron: Asymmetry, 13, 1953–1955. 178 Gutierrez, M.-C., Furstoss, R. and Alphand, V. (2005) Advanced Synthesis and Catalysis, 347, 1051–1059. 179 de Gonzalo, G., Torres Pazmino, D.E., Ottolina, G., Fraaije, M.W. and Carrea, G. (2005) Tetrahedron: Asymmetry, 16, 3077–3083. 180 Alphand, V. and Furstoss, R. (1992) Journal of Organic Chemistry, 57, 1306–1309. 181 Kelly, D.R., Knowles, C.J., Mahdi, J.G., Taylor, I.N. and Wright, M.A. (1995) Journal of the Chemical Society. Chemical Communications, 729–730. 182 Alphand, V., Archelas, A. and Furstoss, R. (1989) Tetrahedron Letters, 30, 3663–3664. 183 Lebreton, J., Alphand, V. and Furstoss, R. (1997) Tetrahedron, 53, 145–160. 184 Mihovilovic, M.D. and Kapitan, P. (2004) Tetrahedron Letters, 45, 2751–2754. 185 Alphand, V. and Furstoss, R. (1992) Tetrahedron: Asymmetry, 3, 379–382. 186 Wang, S., Kayser, M.M. and Jurkauskas, V. (2003) Journal of Organic Chemistry, 68, 6222–6228. 187 Pellissier, H. (2006) Tetrahedron, 62, 5559–5601. 188 Holland, H.L. (1988) Chemical Reviews, 88, 473–485. 189 Colonna, S., Gaggero, N., Richelmi, C. and Pasta, P. (2000) Nato Science Series 1, 33, 133–160. 190 Colonna, S., Gaggero, N., Carrea, G. and Pasta, P. (2001) Asymmetric Oxidation Reactions (ed. T. Katsuki), Oxford University Press, Oxford, pp. 227–235. 191 Colonna, S., Del Sordo, S., Gaggero, N., Carrea, G. and Pasta, P. (2002) Heteroatom Chemistry, 13, 467–473. 192 Holland, H.L., Gu, J.-X., Kerridge, A. and Willetts, A. (1999) Biocatalysis and Biotransformation, 17, 305–317.

j271

j 9 Biooxidations in Chiral Synthesis

272

193 ten Brink, H.B., Holland, H.L., Scoemaker, H.E., van Lingen, H. and Wever, R. (1999) Tetrahedron: Asymmetry, 10, 4563–4572. 194 Andersson, M.A. and Allenmark, S.G. (1998) Tetrahedron, 54, 15293–15304. 195 Fu, H., Kondo, H., Ichikawa, Y., Look, G.C. and Wong, C.-H. (1992) Journal of Organic Chemistry, 57, 7265–7270. 196 Trevisan, V., Signoretto, M., Colonna, S., Pironti, V. and Strukul, G. (2004) Angewandte Chemie International Edition, 43, 4097–4099. 197 Adam, W., Heckel, F., Saha-M€oller, C.R., Taupp, M., Meyer, J.-M. and Schreier, P. (2005) Applied and Environmental Microbiology, 71, 2199–2202. 198 Colonna, S., Gaggero, N., Pasta, P. and Ottolina, G. (1996) Chemical Communications, 2303–2307. 199 Colonna, S., Gaggero, N., Carrea, G. and Pasta, P. (1997) Chemical Communications, 439–440. 200 Kelly, D.T., Knowles, C.J., Mahdi, J.G., Taylor, I.N. and Wright, M.A. (1996) Tetrahedron: Asymmetry, 7, 365–368. 201 de Gonzalo, G., Pazmino, D.E.T., Ottolina, G., Fraaije, M.W. and Carrea, G. (2006) Tetrahedron: Asymmetry, 17, 130–135. 202 de Gonzalo, G., Ottolina, G., Carrea, G. and Fraaije, M.W. (2005) Chemical Communications, 3724–3726. 203 Allen, C.C.R., Boyd, D.R., Dalton, H., Sharma, N.D., Haughey, S.A., McMordie, R.A.S., McMurray, B.T., Sheldrake, G.N. and Sproule, K. (1995) Chemical Communications, 119–120. 204 Zambianchi, F., Pasta, P., Carrea, G., Colonna, S., Gaggero, N. and Woodley, J.M. (2002) Biotechnology and Bioengineering, 78, 489–496. 205 Reetz, M.T., Daligault, F., Brunner, B., Hindrichs, H. and Deege, A. (2004) Angewandte Chemie International Edition, 43, 4078–4081. 206 Holland, H.L., Brown, F.M., Larsen, B.G. and Zabic, M. (1995) Tetrahedron: Asymmetry, 6, 1569–1574.

207 Holland, H.L., Andreana, P.R. and Brown, F.M. (1999) Tetrahedron: Asymmetry, 10, 2833–2843. 208 Olivo, H.F., Osorio-Lozada, A. and Peeples, T.L. (2005) Tetrahedron: Asymmetry, 16, 3507–3511. 209 Colonna, S., Gaggero, N., Carrea, G. and Pasta, P. (1998) Chemical Communications, 415–416. 210 Colonna, S., Gaggero, N., Carrea, G., Pasta, P., Alphand, V. and Furstoss, R. (2001) Chirality, 13, 40–42. 211 Alphand, V., Gaggero, N., Colonna, S. and Furstoss, R. (1996) Tetrahedron Letters, 37, 6117–6120. 212 Alphand, V., Gaggero, N., Colonna, S., Pasta, P. and Furstoss, R. (1997) Tetrahedron, 53, 9695–9706. 213 Colonna, S., Gaggero, N., Carrea, G. and Pasta, P. (1996) Tetrahedron: Asymmetry, 7, 565–570. 214 Kayser, M.M., Zhao, H., Chen, G. and Feicht, A. (2002) Arkivoc, xii, 47–62. 215 Ottolina, G., Bianchi, S., Belloni, B., Carrea, G. and Danieli, B. (1999) Tetrahedron Letters, 40, 8483–8486. 216 Park, S.B., Jacob, P. III, Benowitz, N.L. and Cashman, J.R. (1993) Chemical Research in Toxicology, 6, 880–888. 217 Branchaud, B.P. and Walsh, C.T. (1985) Journal of the American Chemical Society, 107, 2153–2161. 218 Latham, J.A. Jr., Branchaud, B.P. Jr., Chen, Y.-C.J. and Walsh, C. (1986) Journal of the Chemical Society. Chemical Communications, 528–530. 219 Gibson, D.T. and Parales, R.E. (2000) Current Opinion in Biotechnology, 11, 236–243. 220 Boyd, D.R. and Bugg, T.D.H. (2006) Organic and Biomolecular Chemistry, 4, 181–192. 221 Bui, V.P., Hansen, T.V., Stengstrom, Y. and Hudlicky, T. (2000) Green Chemistry, 2, 263–265. 222 Yildirim, S., Zezula, J., Hudlicky, T., Witholt, B. and Schmid, A. (2004) Advanced Synthesis and Catalysis, 346, 933–942.

References 223 Di Gennaro, P., Conforti, P., Lasagni, M., Bestetti, G., Bernasconi, S., Orsini, F. and Sello, G. (2006) Biotechnology and Bioengineering, 93, 511–518. 224 Hudlicky, T., Gonzalez, D. and Gibson, D.T. (1999) Aldrichimica Acta, 32, 35–62. 225 Gibson, D.T., Hensley, M., Yoshioka, H. and Mabry, T.J. (1970) Biochemistry, 9, 1626–1630. 226 Zylstra, G.J. and Gibson, D.T. (1989) Journal of Biological Chemistry, 264, 14940–14946. 227 Jeffrey, A.M., Yeh, H.J.C., Jerina, D.M., Patel, T.R., Davey, J.R. and Gibson, D.T. (1975) Biochemistry, 14, 575–584. 228 Simon, M.J., Osslung, T.D., Saunders, R., Ensley, B.D., Suggs, S., Harcourt, A., Suen, W.C., Cruden, D.L., Gibson, D.T. and Zylstra, G.J. (1993) Gene, 127, 31–37. 229 Di Gennaro, P., Galli, E., Albini, G., Pelizzoni, F., Sello, G. and Gestetti, G. (1997) Research in Microbiology, 148, 355–364. 230 Di Gennaro, P., Sello, G., Bianchi, D. and DAmico, P. (1997) Journal of Biological Chemistry, 272, 30254–30260. 231 Gibson, D.T., Roberts, R.L., Wells, M.C. and Kobal, V.M. (1973) Biochemical and Biophysical Research Communications, 50, 211–219. 232 Mondello, F.J. (1989) Journal of Bacteriology, 171, 1725–1732. 233 Werlen, C., Kohler, H.-P.E. and van der Meer, J.R. (1996) Journal of Biological Chemistry, 271, 4009–4016. 234 Yildirim, S., Zezula, J., Hudlicky, T., Witholt, B. and Schmid, A. (2004) Adv. Synth. Catal., 346, 933–942. 235 Yildirim, S., Franko, T.T., Wohlgemuth, R., Kohler, H.-P.E., Witholt, B. and Schmid, A. (2005) Advanced Synthesis and Catalysis, 347, 1060–1072. 236 Reiner, A.M. and Hegeman, G.D. (1971) Biochemistry, 10, 2530–2536. 237 Knackmuss, H.-J., Beckmann, W. and Otting, W. (1976) Angewandte Chemie, 88, 581–582. 238 Boyd, D.T., Sharma, N.D., Hempenstall, F., Kennedy, M.A., Malone, J.F., Allen,

239

240

241

242 243

244

245

246 247 248

249

250 251

C.C.R., Resnick, S.M. and Gibson, D.T. (1999) Journal of Organic Chemistry, 64, 4005–4011. Boyd, D.R., Sharma, N.D., Ljubez, V., Byrne, B.E., Shepherd, S.D., Allen, C.C.R., Kulakov, L.A., Larkin, M.J. and Dalton H. (2002) Chemical Communications, 1914–1915. Boyd, D.R., Sharma, N.D., Belhocine, T., Malone, J.F., McGregor, S. and Allen, C.C.R. (2006) Chemical Communications, 4934–4936. Boyd, E.R., Sharma, N.D., Gunaratne, N., Haugehy, S.A., Kennedy, M.A., Malone, J.F., Allen, C.C.R. and Dalton, H. (2003) Organic and Biomolecular Chemistry, 1, 984–994. Modyanova, L. and Azerad, R. (2000) Tetrahedron Letters, 41, 3865–3869. Boyd, E.R., Sharma, N.D., Brannigan, I.N. Haugehy, S.A., Malone, J.F., Clarke, D.A. and Dalton, H. (1996) Chemical Communications, 2361–2362. Boyd, D.R., Sharma, N.D., Modyanova, L.V., Carroll, J.G., Malone, J.F., Allen, C.C.R., Hamilton, J.T.G., Gibson, D.T: Parales, R.E. and Dalton H. (2002) Canadian Journal of Chemistry, 80, 589–600. Allen, C.C.R., Boyd, D.R., Dalton, H., Sharma, N.D., Brannigan, I., Kerley, N.A., Sheldrake, G.N. and Taylor, S.C. (1995) Journal of the Chemical Society. Chemical Communications, 117–118. Hudlicky, T. and Thorpe, A.J. (1996) Chemical Communications, 1993–2000. Ley, S. and Sternfeld, F. (1989) Tetrahedron, 45, 3463–3476. Hudlicky, T., Price, J.D., Rulin, F. and Tsunoda, T. (1990) Journal of the American Chemical Society, 112, 9439–9440. Boyd, D.R., Sharma, N.D., Dalton, H. and Clarke, D.A. (1996) Chemical Communications, 45–46. McKibben, B.P., Barnosky, G.S. and Hudlicky, T. (1995) Synlett, 806–808. Franke, D., Lorbach, V., Esser, S., Dose, C., Sprenger, G.A., Halfar, M., Th€ommes, J., M€ uller, R., Takors, F., and M€ uller, M.

j273

j 9 Biooxidations in Chiral Synthesis

274

252

253

254 255

256

257

258

259

260

261

262

263 264

265

(2003) Chemistry- A European Journal, 9, 4188–4196. Hudlicky, T., Restrepo-Sanchez, N., Kary, P.D. and Jaramillo-Gomez, L.M. (2000)CarbohydrateResearch,324,200–203. Hudlicky, T., Rouden, J., Luna, H. and Allen, S. (1994) Journal of the American Chemical Society, 116, 5099–5107. Zezula, J. and Hudlicky, T. (2005) Synlett, 3, 388–405. Boyd, D.R., Sharma, N.D., Llamas, N.M., Malone, J.F., ODowd, C.R. and Allen, C.C.R. (2005) Organic and Biomolecular Chemistry, 3, 1953–1963. Rinner, U., Hillebrenner, H.L., Adams, D.R., Hudlicky, T. and Pettit, G.R. (2004) Bioorganic and Medicinal Chemistry Letters, 14, 2911–2915. Hudlicky, T., Rinner, U., Gonzalez, D., Akgun, H., Schilling, S., Siengalewicz, P., Martinot, T.A. and Pettit, G.R. (2002) Journal of Organic Chemistry, 67, 8726–8743. Austin, K.A.B., Banwell, M.G., Loong, D.T.J., Rae, A.D. and Willis, A.C. (2005) Organic and Biomolecular Chemistry, 3, 1081–1088. Banwell, M.D., Loong, D.T.J. and Willis, A.C. (2005) Australian Journal of Chemistry, 58, 511–516. Newman, L.M., Garcia, H., Hudlicky, T. and Selifonov, S.A. (2004) Tetrahedron, 60, 729–734. Morawski, B., Griengl, H., Ribbons, D.W. and Williams, D.J. (1997) Tetrahedron: Asymmetry, 8, 845–848. Myers, A.G., Siegel, D.R., Buzard, D.J. and Charest, M.G. (2001) Organic Letters, 3, 2923–2926. Parker, M.H., Maryanoff, B.E. and Reitz, A.B. (2004) Synlett, 2095–2098. Jenkins, G.N., Ribbons, D.W., Widdowson, D.A., Slawin, A.M.Z. and Williams, D.J. (1995) Journal of the Chemical Society Perkin Transactions 1, 2647–2655. Mihovilovic, M.D., Leisch, H.G. and Mereiter, K. (2004) Tetrahedron Letters, 45, 7087–7090.

266 Gribble, G.W. (1998) Accounts of Chemical Research, 31, 141–152. 267 van Pee, K.-H. and Unversucht, S. (2003) Chemosphere, 52, 299–312. 268 Gribble, G.W. (2004) Journal of Chemical Education, 81, 1441–1449. 269 Vaillancourt, F.H., Yeh, E., Yosburg, D.A., Garneau-Tsodikova, S. and Walsh, C.T. (2006) Chemical Reviews, 106, 3364–3378. 270 Sundaramoorthy, M., Terner, J. and Poulos, T.L. (1998) Chemistry and Biology, 5, 461–473. 271 Dunford, H.B. (1999) Heme Peroxidases, Wiley-VCH Verlag GmbH, New York pp. 323–348. 272 Taurog, A. and Howells, E.M. (1966) The Journal of Biological Chemistry, 241, 1329–1339. 273 Keller, S., Wage, T., Hohaus, K., H€olzer, M., Eichhorn, E. and van Pee, K.-H. (2000) Angewandte Chemie International Edition, 39, 2300–2302. 274 Butler, A. and Carter-Franklin, J.N. (2004) Natural Products Reports, 21, 180–188. 275 Carter-Franklin, J.N. and Butler, A. (2004) Journal of the American Chemical Society, 126, 15060–15066. 276 Vaillancourt, F.H., Yin, J. and Walsh, C.T. (2005) Proceedings of the National Academy of Sciences of the United States of America, 102, 10111–10116. 277 Blasiak, L.C., Vaillancourt, F.H., Walsh, C.T. and Drennan, C.L. (2006) Nature, 440, 368–371. 278 Vaillancourt, F.H., Vosburg, D.A. and Walsh, C.T. (2006) Chembiochem A European Journal of Chemical Biology, 7, 748–752. 279 Galonic, D.P., Vaillancourt, F.H. and Walsh, C.T. (2006) Journal of the American Chemical Society, 128, 3900–3901. 280 Ueki, M., Galonic, D.P., Vaillancourt, F.H., Garneau-Tsodikova, S., Yeh, E., Vosburg, D.A., Schroeder, F.C., Osada, H. and Walsh, C.T. (2006) Chemistry and Biology, 13, 1183–1191.

j275

10 Aldolases: Enzymes for Making and Breaking C----C Bonds Wolf-Dieter Fessner

10.1 Introduction

Stereoselective carbon–carbon bond construction is a pivotal process in the asymmetric synthesis of complex molecular targets. Enzymes are ﬁnding increasing acceptance as chiral catalysts for the in vitro synthesis of asymmetric compounds because they have been optimized by evolution for high selectivity and catalytic efﬁciency [1–12]. Using enzyme catalysis, molecular complexity can be rapidly built up under mild conditions, without a need for protection of sensitive or reactive functional groups, and with high chemical efﬁciency and uncompromised stereochemical ﬁdelity. Particularly, the high stereospeciﬁcity of aldolases in CC bondforming reactions confers considerable utility as synthetic biocatalysts [13–19], offering an environmentally benign alternative to chiral transition metal catalysis of the asymmetric aldol reaction [20–22]. Several dozens of aldolases have been identiﬁed so far in nature [23,24], and many of these enzymes are commercially available at a scale sufﬁcient for preparative applications. Enzyme catalysis is more attractive for the synthesis and modiﬁcation of biologically relevant classes of organic compounds that are typically complex, multifunctional, and water soluble. Typical examples are those structurally related to amino acids [5–10] or carbohydrates [25–28], which are difﬁcult to prepare and to handle by conventional methods of chemical synthesis and mandate the laborious manipulation of protective groups. Catalytic CC coupling is particularly valuable in asymmetric synthesis because of its potential for stereodivergent product generation [13], by which multiple stereoisomeric products can be derived from common synthetic building blocks (Figure 10.1). Obviously, such a synthetic strategy depends on the prevalence of related stereocomplementary enzymes that must have a similarly broad substrate tolerance.

Asymmetric Organic Synthesis with Enzymes. Edited by Vicente Gotor, Ignacio Alfonso, and Eduardo Garca-Urdiales Copyright 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 978-3-527-31825-4

j 10 Aldolases: Enzymes for Making and Breaking C----C Bonds

276

Figure 10.1 Generation of stereo-diversity by the aldol addition.

10.2 Mechanistic Classification

Most of the enzymes known to facilitate carbon–carbon bond formation and cleavage (lyases) catalyze a crossed aldol reaction as a reversible, stereocontrolled addition of a nucleophilic ketone donor to an electrophilic aldehyde acceptor. From a synthetic perspective, the most useful and most extensively studied enzymes utilize aldol donors comprising 2-C or 3-C fragments and can be grouped into four categories depending on the structure of their nucleophilic component (Figure 10.2): (i) the pyruvate (and phosphoenolpyruvate)–dependent aldolases, (ii) the dihydroxyacetone phosphate(DHAP)-dependent aldolases, (iii) an acetaldehyde-dependent aldolase, and (iv) glycine-dependent enzymes. Members of the ﬁrst and third types generate a-methylene carbonyl compounds and thereby generate a single stereocenter, while members of the other types form a-substituted carbonyl derivatives that contain two new vicinal chiral centers at the new CC bond, a fact that makes them particularly appealing for asymmetric synthesis. Typically, lyases are quite speciﬁc for the nucleophilic donor component owing to mechanistic requirements. Usually, approach of the aldol acceptor to the enzymebound nucleophile occurs stereospeciﬁcally following an overall retention mechanism, while the facial differentiation of the aldehyde carbonyl is responsible for the relative stereoselectivity. In this manner, the stereochemistry of the CC bond formation is completely controlled by the enzymes, in general irrespective of the constitution or conﬁguration of the substrate, which renders the enzymes highly predictable. On the other hand, most of the lyases allow a reasonably broad variation of the electrophilic acceptor component that is usually an aldehyde. This feature O

O H

CO2H HH

Pyruvate

O OPO32–

HO HH

Dihydroxyacetone phosphate

H

O H

HH Acetaldehyde

H2 N

OH HH Glycine

Figure 10.2 Nucleophilic donor substrates of preparatively useful aldolases.

10.2 Mechanistic Classification

j277

Enzyme

H+

O

H2O

H+

NH

X

R

Enzyme

Enzyme

NH2

X

R H+

X

R

A

H2O

NH

B

H+ R'CHO

R'CHO

Enzyme Enzyme

NH2 O

H+

X

R R'

Enzyme

NH H2 O

OH

H+

X

– CO2

R'

H2O

(b)

Zn His HO His His

H+

R'

Enzyme

X

O

X R

O H

OH

CO2H

X

R

C

Enzyme

NH

H+

X

R

(a)

Enzyme

O HO

D

–

Zn

O

CO2H His

O

His His

H

Zn His HO His His

Enzyme—O

Figure 10.3 Schematic mechanism of class I aldolases (a) and of class II aldolases (b).

makes a stereodivergent strategy possible for the synthesis of arrays of stereoisomeric compounds by employing stereocomplementary enzymatic catalysts to selectively produce individual stereoisomers at will from a given starting material. Enzymatic activation of the aldol donor substrates occurs by stereospeciﬁc deprotonation along two different pathways (Figure 10.3) [30,31]. Class I aldolases bind their substrates covalently via imine/enamine formation to an active site lysine residue to initiate bond cleavage or formation [32]. (As a variation, glycine-utilizing aldolases are activating their substrate by the aid of pyridoxal phosphate as an imine forming cofactor.) In contrast, class II aldolases utilize a Lewis acid cofactor to facilitate deprotonation via bidentate coordination of the donor to furnish the enediolate nucleophile [33]. Usually, this effect is achieved by a tightly bound Zn2þ ion but other divalent cations can act in its place [30]. A number of mechanistically distinct enzymes can likewise be employed for the synthesis of product structures identical to those accessible from aldolase catalysis. Such alternative cofactor-dependent enzymes (e.g. transketolase) are emerging as useful catalysts in organic synthesis. As these operations often extend and/or

j 10 Aldolases: Enzymes for Making and Breaking C----C Bonds

278

complement the synthetic strategies open to aldolases, a selection of such enzymes and examples of their synthetic utility are included in this overview.

10.3 Pyruvate Aldolases

Pyruvate-dependent lyases serve catabolic functions in vivo in the degradation of sialic acids and KDO (2-keto-3-deoxy-manno-octosonate), and in that of 2-keto-3deoxy aldonic acid intermediates from hexose or pentose catabolism. As these freely reversible aldol additions often have less favorable equilibrium constants [30,34], synthetic reactions usually have to be driven by an excess of pyruvate to achieve satisfactory conversions. A few related enzymes have been identiﬁed that utilize phosphoenolpyruvate instead of pyruvate, which upon CC bond formation releases inorganic phosphate, and thus renders the aldol addition essentially irreversible (Figure 10.4) [16]. Although attractive from a synthetic point of view, the latter enzymes have been less studied as yet for preparative applications [35]. 10.3.1 N-Acetylneuraminic Acid Aldolase

N-Acetylneuraminic acid aldolase (or sialic acid aldolase, NeuA; EC 4.1.3.3) catalyzes the reversible addition of pyruvate (2) to N-acetyl-D-mannosamine (ManNAc; (1)) in the degradation of the parent sialic acid (3) (Figure 10.4). The NeuA lyases found in both bacteria and animals are type I enzymes that form a Schiff base/enamine intermediate with pyruvate and promote a si-face attack to the aldehyde carbonyl group with formation of a (4S) conﬁgured stereocenter. The enzyme is commercially available and it has a broad pH optimum around 7.5 and useful stability in solution at ambient temperature [36]. O +

HO HO

OH NHAc O

OH

2 OH

1

NeuA CO2H

OH

CO2H Phosphoenol pyruvate

O CO2H

S

OPO32– +

OH

HO OH

NHAc

NeuS Pi HO

OH

AcHN HO OH

Figure 10.4 Natural substrates of the N-acetylneuraminic acid aldolase (NeuA) and synthase (NeuS).

OH O

3

CO2H

10.3 Pyruvate Aldolases

OH

HO O

HO HO

4

OH NHAc

CO2H

O

HO OH AcHN

NH2

HN Zanamivir

NH Epimerase Steps HO HO HO

NeuA

NHAc O OH

1

Pyruvate

HO

OH

AcHN HO OH

OH O

CO2H

3

Figure 10.5 Industrial process for the production of N-acetylneuraminic acid as a precursor to an influenza inhibitor.

The natural substrate of the aldolase has been an important target for synthesis. Zanamivir was introduced in 1999 for the treatment of infection caused by inﬂuenza and represented the ﬁrst medicinally used sialic acid derivative with antiviral activity. As the precursor (3) is a rare and expensive natural product, this required a cost-effective synthesis. Thus, NeuA was the ﬁrst aldolase to ﬁnd attention for the development of an industrial bioconversion process that is now operated at multiton scale (Figure 10.5) [37–40]. In this equilibrium-controlled bioconversion (equilibrium constant 12.7 M1 in favor of the retroaldolization), the expensive (1) can be produced by an integrated enzymatic in situ isomerization from inexpensive N-acetylglucosamine (4) by a combination of NeuA with an N-acylglucosamine 2-epimerase (EC 5.1.3.8) catalyst in an enzyme membrane reactor (EMR) [41–43]. Product isolation, however, is complicated by the excess of pyruvate needed to achieve a preparative useful conversion [39]. Very recently, efforts have been directed toward evolution of NeuA mutants that tolerate substrate modiﬁcations to facilitate a shortened synthesis of marketable neuraminidase inhibitors [44,45]. Another approach utilized a coupling of bacteria expressing the 2-epimerase activity and NeuS activities independently [46]. As an example for continuous process design, 2-keto-3-deoxy-D-glycero-D-galactononosouate (KDN) (5) has been produced on a 100-g scale from D-mannose and pyruvate using a pilot-scale EMR at a space-time yield of 375 g l1 d1 and an overall crystallized yield of 75% (Figure 10.6) [47]. Similarly, L-KDO (6) can be synthesized from L-arabinose [48]. Extensive studies have indicated that only pyruvate is acceptable as the NeuA donor substrate, with the exception of ﬂuoropyruvate [49], but that the enzyme displays a fairly broad tolerance for stereochemically related aldehyde substrates as acceptor alternatives, such as a number of sugars and their derivatives larger or equal to pentoses [36,48,50,51]. Permissible variations include replacement of the natural D-manno conﬁgured substrate (4) with derivatives containing modiﬁcations such as epimerization, substitution, or deletion at positions C-2, -4, or -6 [16,27]. Epimerization at C-2, however, is restricted to small polar substituents owing to strongly

j279

j 10 Aldolases: Enzymes for Making and Breaking C----C Bonds

280

N-Acetyl-D- NeuA mannosPyruvate amine

OH

OH

HO

OH

NHAc S HO OH

O

CO2H

3

Pyruvate

O

HO

(a)

D-KDN

O

L-Arabinose

O CO2H

R OH ent-5 L-KDN

OH

HO

Pyruvate

OH

HO HO HO

CO2H

HO

Pyruvate

N-Acetyl-Lmannosamine

NeuA

OH

5

NeuA

CO2H

R OH ent-3 L-NeuAc

D-NeuAc

OH

HO S HO OH

D-Mannose

O

OH

HO NeuA

OH

HO NHAc HO

HO

CO2H

6

L-KDO

L-Mannose

Pyruvate

HO OH O

HO

CO2H R

OH

NeuA

OH (b)ent-6 D -KDO

NeuA

D-Arabinose

Pyruvate

Figure 10.6 Sialic acids prepared on larger scale (a), and unusual products formed with inverted (4R)-configuration (b).

decreasing reaction rates [52,53]. The broad substrate tolerance of the catalyst for sugar precursors has recently been exploited in the equilibrium generation of sialic acid and analogs for an in situ screening of a dynamic combinatorial library [54,55]. On account of the importance of sialic acids in a wide range of biological recognition events, the aldolase has become instrumental for the chemoenzymatic synthesis of a multitude of other natural and unnatural derivatives or analogs of (3) (Figure 10.7). A large number of examples have been reported for sialic acid modiﬁcations at C-5/C-9 [16,27] such as differently N-acylated derivatives (8) [56–59], including amino acid conjugates [60], or 9-modiﬁed analogs (9) [61–63] from suitable mannosamine precursors (7) in the search for new neuraminidase (inﬂuenza) inhibitors. Most notably, the N-acetyl group in (1) may either be omitted [52,53] or replaced by sterically

X

NHCOR O OH

HO HO

OH

HO

NeuA Pyruvate

7

X

OH

RCOHN HO OH

O

CO2H

OH OH

AcHN HO OH

8

O

CO2H

9

NHBoc R = OtBu, CH2OH,

O O

CH2Ph

X = CH2OCH3, CH2OMOM, CH2OAc, CH2OBz

NH O O

OH

AcHN HO OH

HO

OH O

CO2H Bn

10

HO

OH OH H O N HO OH

O O

CO2H

11

Figure 10.7 Neuraminic acid derivatives accessible by NeuA catalysis, including an intermediate for alkaloid synthesis.

OH

HO N

HO

12

OH

10.3 Pyruvate Aldolases

Figure 10.8 Three-point binding model for prediction of NeuA stereoselectivity based on conformational analysis.

demanding substituents such as N-Cbz (11) [64,65] or even a nonpolar phenyl group [52] without destroying activity. Large acyl substituents are also tolerated at C-6 as shown by the conversion of a Boc-glycyl derivative (10) as a precursor to a ﬂuorescent sialic acid conjugate [66], as well as by C-6 derivatives containing a perﬂuorohexyl-tag to serve as a puriﬁcation aid [67]. In most of the cases investigated so far, a high level of asymmetric induction by NeuA for the (4S) conﬁguration is retained. However, some carbohydrates were also found to be converted with random or even inverse stereoselectivity for the C-4 conﬁguration, such as for ent-3, ent-5, or ent-6 (Figure 10.6) [48,51,68–71]. A critical and distinctive factor seems to be recognition of the conﬁguration by the enzymic catalyst at C-3 in the aldehydic substrate [48,51]. A three-point binding model has been proposed to account for the inverse conformational preference in direction of synthesis (Figure 10.8) [13]. On the basis of the (3S)-a-4 C1 structure of the natural substrate and a conformational analysis of its analogs, this model can predict the occasionally observed compromise in, or even total inversion of, the facial stereoselectivity of CC bond formation. Starting from the N-Cbz-protected aldolase product (11), azasugar (12) has been obtained stereoselectively by intramolecular reductive amination as an analog of the bicyclic, indolizidine-type glycosidase inhibitor castanospermine (Figure 10.7) [65]. Also, it had been recognized that the C-12–C-20 sequence of the macrolide antibiotic amphotericin B resembles the b-pyranose tautomer of (14) (Figure 10.9). Thus, the branched-chain manno-conﬁgured substrate (13) was successfully chain-extended under NeuA catalysis to yield the potential amphotericin B synthon (14) in good yield [72,73]. 10.3.2 Related Pyruvate Aldolases

The KDO aldolase (KdoA, EC 4.1.2.23) is involved in the catabolism of the eight-carbon sugar D-KDO, which is reversibly degraded to D-arabinose (15) and pyruvate (Figure 10.10). The enzyme has been partially puriﬁed from bacterial sources and studied for synthetic applications [71,74]. It seems that the KdoA, similar to

j281

j 10 Aldolases: Enzymes for Making and Breaking C----C Bonds

282

HO HO HO

OH

HO

OH NeuA

O OH

O

Pyruvate

HO2C

OH CO2H

O

14

OH OH

12 OH

O OH

OH

HO

OH

O

OH

HO HO OH

13

HO

OH

OH

OH

OH

O

OH 16

CO2H

20

OR

Amphotericin B

Figure 10.9 NeuA catalyzed preparation of a synthetic precursor to the macrolide antibiotic amphotericin B.

NeuA, has broad substrate speciﬁcity for aldoses while pyruvate was found to be irreplaceable. As a notable distinction, KdoA was also active on smaller acceptors such as glyceraldehyde. Preparative applications, for example, for the synthesis of KDO (ent-6) and its homologs or analogs (16)/(17), suffer from an unfavorable equilibrium constant of 13 M1 in direction of synthesis [34]. The stereochemical course of aldol additions generally seems to adhere to a re-face attack on the aldehyde carbonyl, which is complementary to the stereoselectivity of NeuA. On the basis of the results published so far, it may be concluded that a (3R)-conﬁguration is necessary (but not sufﬁcient), and that stereochemical requirements at C-2 are less stringent [71]. A class I aldolase speciﬁc for cleavage of 2-keto-3-deoxy-6-phospho-D-gluconate (19) (KDPGlc aldolase or KdgA; EC 4.1.2.14) is produced by many microorganisms for the degradation of 6-phosphogluconate to give pyruvate and D-glyceraldehyde 3-phosphate (18) (Figure 10.11). The equilibrium constant favors synthesis (103 M1) [75].

O OH OH

O OH

HO

+

HO

HO

CO2H

15

OH

O CO2H

R

OH

OH

2

OH OH

OH

KdoA

HO

OH

HO

O

HO

CO2H

16 OH

F

HO

O

HO

CO2H

17 OH

Figure 10.10 Natural substrates of the 2-keto-3-deoxy-mannooctosonic acid aldolase, and nonnatural sialic acids obtained by KdoA catalysis.

HO OH O

HO

CO2H ent - 6

OH

10.3 Pyruvate Aldolases KDPGlc aldolase

OH 2– O PO 3

2–

O3PO

CO2H

S

OH

O

CO2H

S

OH

O

HO

19

+ 2

H

OH

OH

18

2–O PO 3

O

j283

2– O

KDPGal aldolase

3PO

2– O

O

3PO

CO2H

R

OH

O

HO

CO2H

R

OH

20

Figure 10.11 Aldol reactions catalyzed in vivo by the 2-keto-3deoxy-6-phospho-D-gluconate and 2-keto-3-deoxy-6-phospho-Dgalactonate aldolases.

Enzyme preparations from liver or microbial sources were reported to show rather high substrate speciﬁcity [76] for the natural phosphorylated acceptor D-(18) but, at much reduced reaction rates, offer a rather broad substrate tolerance for polar, shortchain aldehydes [77–79]. Simple aliphatic or aromatic aldehydes are not converted. Therefore, the aldolase from Escherichia coli has been mutated for improved acceptance of nonphosphorylated and enantiomeric substrates toward facilitated enzymatic syntheses of both D- and L-sugars [80,81]. High stereoselectivity of the wild-type enzyme has been utilized in the preparation of compounds (23)/(24) and in a two-step enzymatic synthesis of (22), the N-terminal amino acid portion of nikkomycin antibiotics (Figure 10.12) [82]. Comparable to the situation for the NeuA and KdoA enzyme pair (see above), a class I lyase complementary to the KDPGlc aldolase that has a stereopreference for the (4S)-conﬁguration is known (Figure 10.11). The aldolase, which acts on

KDPGlc aldolase H

N

NH3

Pyruvate

O

21

CO2H

N OH

O

NADH

CO2H

N

PheDH NAD+

22

OH

NH2

ee > 99.7% CO2

Formate DH HCO3– O O

OH O

CO2H

S

HO

OH

23

H3C

O

S

HO

OH

N OH

NH2

NH N H

O

N

CO2H

24

Nikkomycin Kz

Figure 10.12 Stereoselective synthesis of the amino acid portion of nikkomycin antibiotics and hexulosonic acids using KDPGlc aldolase.

HO

CH3

O

j 10 Aldolases: Enzymes for Making and Breaking C----C Bonds

284

2-keto-3-deoxy-6-phospho-D-galactonate (20) (KDPGal aldolase; EC 4.1.2.21), was also studied for synthetic applications [83].

10.4 Dihydroxyacetone Phosphate Aldolases

Whereas pyruvate aldolases form only a single stereogenic center, aldolases speciﬁc for (DHAP, (25)) as a nucleophile create two new asymmetric centers at the termini of the new CC bond. The fact that nature has evolved a full set of four unique aldolases (Figure 10.13) to cleave all possible stereochemical permutations of the vicinal diol at C-3/C-4 of ketose 1-phosphates (26)–(29) during the retro-aldol cleavage [16,29] is particularly useful for synthetic applications. These aldolases have proved to be exceptionally powerful tools for asymmetric synthesis, particularly for the stereocontrolled synthesis of polyoxygenated compounds, because of their relaxed substrate speciﬁcity, high level of stereocontrol, and commercial availability. In the direction of synthesis, this situation formally allows to generate all four possible stereoisomers of a desired product in a building block fashion [16]. In this manner, the deliberate preparation of a speciﬁc target molecule can be addressed by simply choosing the corresponding enzyme and suitable starting material, thereby offering full control over constitution as well as absolute and relative conﬁguration of the desired

O X

O

H

OPO32–

OH

25

OH

2–

O3PO

OH

O OPO3

HO

HO

2–

FruA

OH

RhuA

3S,4R 3R,4S

O3PO

OH

O OPO3

HO

HO

27

L-Rhamnulose 1-phosphate

2–

HO

OPO32– HO

26

D-Fructose 1,6-bisphosphate

2–

O

H3C

TagA

FucA

OH

3S,4S 3R,4R

28

D-Tagatose 1,6-bisphosphate

Figure 10.13 Aldol reactions catalyzed in vivo by the four stereocomplementary dihydroxyacetone phosphate-dependent aldolases.

O OPO32–

H3 C HO

HO

29

L-Fuculose 1-phosphate

10.4 Dihydroxyacetone Phosphate Aldolases 2–

O3PO

O HO HO

2–

O3PO

OH

FruA OPO32–

O O

3PO

26

O HO

+

OH

FSA

25

O

OH 2–

O

O3PO

30

18

OPO32–

HO

18

OH HO

OH 2–O

j285

+

HO

OH

31

Figure 10.14 Natural glycolytic substrates of the fructose 1,6-bisphosphate aldolase (FruA) and fructose 6-phosphate aldolase (FSA).

product. All DHAP aldolases are quite speciﬁc for the phosphorylated nucleophile (25), which therefore must be prepared independently or generated in situ. 10.4.1 Fructose 1,6-Bisphosphate Aldolase

The D-fructose 1,6-bisphosphate aldolase (FruA; EC 4.1.2.13) catalyzes in vivo the equilibrium addition of (25) to D-glyceraldehyde 3-phosphate (GA3P, (18)) to give D-fructose 1,6-bisphosphate (26) (Figure 10.14). The equilibrium constant for this reaction of 104 M1 strongly favors synthesis [34]. The enzyme occurs ubiquitously and has been isolated from various prokaryotic and eukaryotic sources, both as class I and class II forms [30]. Typically, class I FruA enzymes are tetrameric, while the class II FruA are dimers. As a rule, the microbial class II aldolases are much more stable in solution (half-lives of several weeks to months) than their mammalian counterparts of class I (few days) [84–86]. The class I FruA isolated from rabbit muscle aldolase (RAMA) is the aldolase employed for preparative synthesis in the widest sense, owing to its commercial availability and useful speciﬁc activity of 20 U mg1. Its operative stability in solution is limiting, but the more robust homologous enzyme from Staphylococcus carnosus has been cloned for overexpression [87], which offers unusual stability for synthetic purposes. Recently, it was shown that less polar substrates may be converted as highly concentrated water-in-oil emulsions [88]. Literally hundreds of aldehydes have so far been tested successfully by enzymatic assay and preparative experiments as a replacement for (18) in rabbit muscle FruA catalyzed aldol additions [16,25], and most of the corresponding aldol products have been isolated and characterized. The rabbit FruA can discriminate racemic DL-(18), its natural substrate, with high preference for the D-antipode, but kinetic enantioselectivity for nonionic aldehydes is rather low [84,89]. Functionally related to FruA is the novel class I fructose 6-phosphate aldolase (FSA) from E. coli, which catalyzes the reversible cleavage of fructose 6-phosphate (30) to give dihydroxyacetone (31) and D-(18) [90]. It is the only known enzyme that does not require the expensive phosphorylated nucleophile DHAP for synthetic purpose.

j 10 Aldolases: Enzymes for Making and Breaking C----C Bonds

286

Using (31) as the nucleophile, FSA has been shown to accept several aldehydes as acceptor components for preparative synthesis [91]. In addition to (31), it also utilizes hydroxyacetone as an alternative donor to generate 1-deoxysugars such as (66) regioselectively (Figure 10.25). 10.4.2 Related DHAP Aldolases

The D-tagatose 1,6-bisphosphate aldolase (TagA; EC 4.1.2.40) is involved in the catabolism of D-galacto-conﬁgured carbohydrates and catalyzes the reversible cleavage of D-tagatose 1,6-bisphosphate (28) to D-(18) and (25) (Figure 10.13). Enzymes of class I seem to have apparently no stereochemical selectivity with regard to distinction of (26)/(28) [92], while class II aldolases show a high stereoselectivity for the natural substrate in both cleavage and synthesis directions [93,94]. Utilizing the synthetic capacity of a TagA puriﬁed from E. coli the all-cis (3S, 4S)conﬁgured D-tagatose 1,6-bisphosphate (26) has been prepared from (31) by an expeditious multienzymatic system (Figure 10.15) [93,94]. The aldolase also accepts a range of unphosphorylated aldehydes as substrates but produces diastereomeric mixtures only. This lack of stereoselectivity with generic substrate analogs, which makes native TagA enzymes synthetically less useful, has recently stimulated protein engineering studies to improve its properties [95]. The L-fuculose 1-phosphate aldolase (FucA; EC 4.1.2.17) and the L-rhamnulose 1-phosphate aldolase (RhuA; EC 4.1.2.19) are found in many microorganisms where they are responsible for the degradation of deoxysugars L-fucose and L-rhamnose to give (25) and L-lactaldehyde (Figure 10.16). FucA is speciﬁc for cleavage and synthesis of a D-erythro diol unit while RhuA recognizes the corresponding L-threo conﬁguration. Both enzymes are active as Zn2þ-dependent homotetramers [86]. Like a number of other aldolases, both the RhuA and FucA enzymes are commercially available. Overall practical features make the FucA and RhuA enzymes quite similar for synthetic applications. Both metalloproteins are quite robust under conditions of organic synthesis and show a very high stability in the presence of low Zn2þ concentrations with half-lives in the range of months at room temperature. The

O HO

OH

Glycerol kinase

O OPO32–

HO

31

25 ATP

2– O

ADP

TagA

TPI Pyruvate Pyruvate kinase PEP

OPO3 H

HO

O HO

28

OH O

3PO

2–

18

Figure 10.15 Enzymatic one-pot synthesis of D-tagatose 1,6-bisphosphate based on the stereoselective TagA from E. coli.

OH OPO32–

10.4 Dihydroxyacetone Phosphate Aldolases OPO32–

O

HO

RhuA DHAP

OH

H 3C

FucA DHAP

O H3C

H

OH

OPO32–

O

OH

H 3C HO

OH

OH 29

27

Figure 10.16 Natural substrates of microbial deoxysugar phosphate aldolases.

enzymes even tolerate the presence of large fractions of organic cosolvents (30%) [86] and they are active in highly concentrated water-in-oil emulsion systems [96,97]. Both offer a very broad substrate tolerance for variously substituted aldehydes, which is very similar to that of the FruA enzymes. Characteristically, the RhuA has the greatest tolerance for sterically congested acceptor substrates, as exempliﬁed in the conversion of the tertiary aldehyde 2,2-dimethyl-3-hydroxypropanal [16]. The stereospeciﬁcity of both enzymes for an absolute (3R)-conﬁguration is mechanism-based (see above). FucA generally directs an attack of the DHAP enolate to the si-face of an approaching aldehyde carbonyl and is thereby speciﬁc for synthesis of a (3R,4R)-cis diol unit [86,98], while RhuA controls a re-face attack to create the corresponding (3R,4S)-trans conﬁguration [86]. However, this speciﬁcity for a vicinal conﬁguration is somewhat substrate-dependent, in that simple aliphatic aldehydes can give rise to a certain fraction of the opposite diastereomer [16,86,97]. Stereocontrol, in general, is usually highly effective with aldehydes carrying a 2- or 3hydroxyl group. In addition, both aldolases offer a powerful kinetic preference for L-conﬁgured enantiomers of 2-hydroxyaldehydes (32) (Figure 10.17), which facilitates racemate resolutions [99,100]. Essentially, this feature allows the concurrent determination of three contiguous chiral centers in ﬁnal products (33) or (34) having an L-conﬁguration (de 95) even when starting from the more readily accessible racemic material.

RhuA

OH

O +

R

de ≥ 90%

33

DHAP

OH

OH

R HO

D-32

OPO32–

O

FucA

O +

R

H OH

OH

34

H OH

OH H +

D,L-32

OPO32–

O

R

O R

HO

de ≥ 90%

D-32

R = H3C–, H5C2–, H2C=CH–, H2C=CH–CH2–, FH2C–, N3CH2–, H3COCH2– Figure 10.17 Kinetic enantiopreference of class II DHAP aldolases useful for racemic resolution of a-hydroxyaldehydes.

j287

j 10 Aldolases: Enzymes for Making and Breaking C----C Bonds

288

10.4.3 Preparative Applications

Apparently, all DHAP aldolases are highly speciﬁc for (25) as the donor component for mechanistic reason [30–33], a fact which requires an economical access to this compound for synthetic applications. Owing to the limited stability of (25) in solution, particularly at alkaline pH, it is preferentially generated in situ to avoid high stationary concentrations. The most convenient method is the formation of two equivalents of (25) by retroaldol cleavage from commercially available (26) by the combined action of FruA and triose phosphate isomerase (Figure 10.18 inset) [84]. This scheme has been extended into a highly integrated, artiﬁcial metabolism for the efﬁcacious in situ preparation of (25) from inexpensive feedstock such as glucose and fructose (two equivalents of OH O

HO HO

OH

HO OH O Sucrose HOH2C

HO O

HO HO

CH2OH

OH

O

Glucose OH

O OH HO

OH

OH OH

Fructose

5 Enzymes ATP, PEP (a) 2–O

3PO

O OH HO

D-threo

HO

OPO32–

26 FruA eco O

OH OPO32–

O H (b)

18

TPI

O OPO32–

HO

25

OH

R FucA

O OPO32–

R OH

(c) Figure 10.18 Enzymatic in situ generation of dihydroxyacetone phosphate from fructose 1,6-bisphosphate (b), with extension to an in vitro artificial metabolism for its preparation from inexpensive sugars along the glycolysis cascade (a), and utilization for subsequent stereoselective carbon–carbon bond formation using an aldolase with distinct stereoselectivity (c).

D-erythro

10.4 Dihydroxyacetone Phosphate Aldolases

OH OPO32–

HO

+

35

OH OPO32–

HO

GPO pH 7.5

O HO

O2 1/2

OPO3

2–

FruA pH 7.5

OH

j289

O OPO32–

Butanal

H2O2

OH

25

H2O

(a) Pi

Phytase pH 4.0

Phytase pH 4.0

Pi

PPi OH

OH

(b)

HO

O OH

OH 4 Steps, 1 Pot Glycerol

OH

36

Figure 10.19 Oxidative enzymatic generation of dihydroxyacetone phosphate in situ for stereoselective aldol reactions using DHAP aldolases (a), and extension by pHcontrolled, integrated precursor preparation and product liberation (b).

(25) each), or sucrose (four equivalents of (25)) by a combination of up to seven inexpensive enzymes in vitro (Figure 10.18) [101]. When employing the class II FruA of E. coli for aldol cleavage, which displays high substrate speciﬁcity for (18) and is thus inactive with other added aldehyde substrates, this system can be metabolically engineered by adding another aldolase to yield products having a different stereospeciﬁcity [13]. An advanced technique for the clean generation of (25) in situ is based on the oxidation of L-glycerol 3-phosphate (35) catalyzed by microbial ﬂavine-dependent glycerol phosphate oxidases (GPO) (Figure 10.19, box) [102]. This method generates (26) practically quantitatively and with high chemical purity without a need for separate cofactor regeneration. Both oxygen from air and from a H2O2/catalase system can be used to sustain oxygenation [102,103]. Since DHAP aldolases were found to be insensitive to oxygenated solutions, the oxidative generation of (25) can be smoothly coupled to synthetic aldol reactions [102]. Recently, this method has been extended to include a reversible glycerol phosphorylation by phytase, an inexpensive acid phosphatase, from inexpensive pyrophosphate [104], or controlled ring opening of glycidol by inorganic phosphate [105]. Furthermore, the GPO procedure can also be used for a preparative synthesis of the corresponding phosphorothioate (37), phosphoramidate (38), and methylene phosphonate (39) analogs of (25) (Figure 10.20) from suitable diol precursors [106] to be used as aldolase substrates [102]. In fact, such isosteric replacements of the phosphate ester oxygen were found to be tolerable by a number of class I and class II aldolases, and only some speciﬁc enzymes failed to accept the less polar phosphonate (39) [107]. Thus, sugar phosphonates (e.g. (71)/(72)) that mimic metabolic intermediates but are hydrolytically stable to phosphatase degradation can be rapidly synthesized (Figure 10.28).

j 10 Aldolases: Enzymes for Making and Breaking C----C Bonds

290

O HO

O O

PO3

2–

25

(a)

HO

O S

PO3

2–

HO

37

H N

O PO3

38

2–

HO

PO32–

39

O HO

OH O

31

HO

HX

OX

40 X = AsO32–

Ketose

(b)

Ketose X-ester

DHAP aldolase Aldehyde

VO42–

Figure 10.20 Substrate analogs of dihydroxyacetone phosphate accessible by the GPO oxidation method, and spontaneous, reversible formation of arsenate or vanadate analogs of dihydroxyacetone phosphate in situ for enzymatic aldol additions.

Interestingly enough, dihydroxyacetone (31) in the presence of higher concentrations of inorganic arsenate reversibly reacts to form the corresponding arsenate ester (40) in situ, which can replace (25) as a donor in enzyme-catalyzed aldol reactions (Figure 10.20) [108,109]. However, this procedure suffers from rather low reaction rates and the high toxicity of arsenates. Inorganic vanadate also spontaneously forms the corresponding vanadate ester analog under conditions that reduce its oxidation potential, but so far only RhuA could be shown to accept the vanadate mimic (40) for preparative conversions [13]. Typical applications of the DHAP aldolases concern the synthesis of monosaccharides and derivatives of sugars from suitable functionalized aldehyde precursors. Complex 8- and 9-C monosaccharide derivatives (such as (41); Figure 10.21) could be obtained from pentose and hexose monophosphates by stereospeciﬁc chain extension using FruA from rabbit muscle [110]. High conversion rates and yields are generally achieved with 2- or 3-hydroxyaldehydes because in such cases reaction equilibria proﬁt from the fact that, in aqueous solution, the products will cyclize to give more stable furanose or pyranose isomers. For example, enantiomers of glyceraldehyde are good substrates, and stereoselective addition of (25) produces enantiopure ketohexose 1-phosphates in high yield [84,86,99], from which the free ketosugars are obtained by enzymatic dephosphorylation. For example, the less common L-conﬁgured L-fructose and L-tagatose can be prepared directly from racemic glyceraldehyde by RhuA or FucA catalysis making use of the high kinetic enantioselectivity of Zn2þ-dependent aldolases (Figure 10.21) [99,111]. The general approach has been followed for the de novo synthesis of a multitude of differently substituted, unsaturated [112,113] or regiospeciﬁcally labeled sugars [102,114]. Unusual branched-chain (42), (43)) and spiro-annulated sugars (45), (46)) have been synthesized from the corresponding aldehyde precursors

10.4 Dihydroxyacetone Phosphate Aldolases OH

HO

O OPO32–

HO RhuA

O HO

OH

P'ase

O HO

H

OH

FucA

OH

O OPO32–

HO

DL-32

P'ase

HO OH L-Tagatose

OH

(a)

OH 2–

OH

O3PO

O HO

OH OH

(CH3)2N

(b)

47

HO OH

O OCH3 HO OH HO

HO HO

O

H3C

41

H N SO2

HO

OH

OPO32–

OH

H3C

OPO32–

OH

HO

OPO32–

43

OH

O O

OH CH3

45

OH

O HO

H3C

42

O HO HO

44

OH

O

HO OH

OH

HO OH L-Fructose

OH

DHAP

OH

j291

HO

OPO32–

OH

46 N

OH

OH O

HO

O

F17C8 OH

HO

N OH

OH

OH

48

HO

O HO

HO HO

NH2 N

N

49

Figure 10.21 Aldolase-catalyzed asymmetric synthesis of uncommon L-configured sugars (a), and selected examples of carbohydrate-related product structures that are accessible by enzymatic aldolization (b).

(Figure 10.21) [101,115]. 6-Substituted D-fructofuranoside derivatives such as aromatic sulfonamide (44) (a low nanomolar Trypanosoma brucei inhibitor) [116] are accessible via 6-azido-6-deoxyfructose from 3-azido-2(R)-hydroxypropanal (75) by FruA catalysis [85,117]. In an approach resembling the inversion strategy (see below) an a-C-mannoside (47) has been prepared from D-ribose 5-phosphate [118]. The synthesis of 6-C-perﬂuoroalkyl-D-fructose (48) met the challenges from the strong hydrophobicity and electron withdrawing capacity of a ﬂuorous chain, as well as the products potential surfactant properties [119]. The L-sorbo-conﬁgured homo-C-nucleoside analog (49) has been synthesized as a structural analog to adenosine from an enantiopure (S)-aldehyde precursor [120]. On the basis of FruA catalyzed aldol reactions, 3-deoxy-D-arabino-heptulosonic acid 7-phosphate (DAHP) (52), an intermediate of the shikimic acid pathway, has been synthesized from N-acetylaspartic semialdehyde (50) (Figure 10.22) [121]. Precursors to KDO (see Section 10.3.2) and its 4-deoxy analog have been prepared by FruA catalysis from aldehydes (53) that incorporate an acrylic moiety for further functionalization [122].

j 10 Aldolases: Enzymes for Making and Breaking C----C Bonds

292

H O

=

O3PO

FruA DHAP

CO2CH3

OPO3= COOH

(AcO)3BH–

NHAc

OH

OH

OH

AcHN

51

50

FruA DHAP

X H

OH

OH

O

HO HO

CO2H

52 DAHP HO HO

X

OH

OH O CO2H

X P'ase

O

O

COOH

HO

OH

COOH

X = OH KDO X=H 4-Deoxy-KDO

54

53

Figure 10.22 Synthetic approaches to DAHP and KDO by a backbone inversion strategy using FruA catalysis.

HO O 2–O

OH

OH

OH O

OH

O

O

OPO32–

OH

3PO

O

O OH

55

56

OH

OH

O OPO32–

HO

GPO

OPO32–

HO

2–O

3PO

OH

35 O2 HO

Cat

OCH3 OH

GalO

HO

H2O2

O

RhuA HO

OH O

HO

25

OH O HO

CHO O

OCH3 OH

OCH3

HO OH

OH

57

HO

2– O

HO

O

OH

O

3PO

58

OCH3

OH

Figure 10.23 Sialyl Lewisx-related selectin inhibitor and fluorogenic screening compound for transketolase prepared using enzymatic aldolization, and multienzymatic oxidation–aldolization strategy for the synthesis of bicyclic higher carbon sugars.

Fluorogenic compound (56) for transketolase assays has been prepared making use of FruA speciﬁcity [123]. Pendant anionically charged chains have been extended from O- or C-glycosidic aldehydes to furnish low molecular weight mimics of the sialyl Lewis X tetrasaccharide such as (55) (Figure 10.23) [124]. Other higher carbon

10.4 Dihydroxyacetone Phosphate Aldolases

j293

sugar derivatives such as the bicyclic sugar (58) have been prepared by diastereoselective chain extension of simple alkyl galactosides (57) after their terminal oxidation in situ by using a galactose oxidase (GalO; EC 1.1.3.9). The whole scheme can be conveniently effected as a one-pot operation including the parallel generation of (25) by the GPO method [125]. Further bicyclic carbohydrate structures similar to (58) have also been achieved by uni- [87] and bidirectional extension of dialdehyde substrates (Figure 10.32) [126]. Class II aldolases are effective in the kinetic resolution of racemic 2-hydroxyaldehydes (Figure 10.17). Under fully equilibrating conditions, however, diastereoselectivity of aldolase reactions can be steered by thermodynamic control to favor the energetically most stable product [84,101,127,128]. Particularly strong discrimination results from utilization of 3-hydroxylated aldehydes such as (59) owing to the cyclization of products in water to form a pyranoid ring (Figure 10.24). The pronounced conformational destabilization by diaxial repulsions (61) strongly supports those diastereoisomers having a maximum of equatorial substituents [126,128]. Thus, in FruA catalyzed reactions (3S)-conﬁgured hydroxyaldehydes are the preferred substrates to give a most stable all-equatorial substitution in the product (such as (60)) with up to 95% de. Similarly, 2-alkylated aldehydes can be resolved because of the high steric preference of an alkyl group for an equatorial position [127]. Owing to the enantio-complementary nature of the FruA–RhuA biocatalyst pair, under conditions of thermodynamic control this enables to construct mirror imaged products (60) (FruA) and ent-(60) (RhuA) from racemic 3-hydroxybutanal (59) with similar selectivity, but with preference for opposite enantiomers [13]. The all-equatorial substitution in the predominant product can facilitate its separation by crystallization so that the remaining mixture may be resubjected to further equilibration in order to

OH

OH

OH

O

OPO3=

S

H3C

H 3C OH

O

H 3C

OH H

OPO3=

OH

OH

O

R

OH

X 3C

H

60 OH

H 3C

H3C

O

OH

HO

FruA

59 + DHAP

OH

O

OPO3=

O

OPO3=

OH

61

HO

HO

1. FBP, FruA, TPI 2. P'ase X3C

O

OH OH

OH

NaIO4 NaBH4

62 X = H, F Figure 10.24 Diastereoselectivity in FruA catalyzed aldol additions to 3hydroxyaldehydes under thermodynamic control, and synthesis of L-fucose derivatives based on thermodynamic preference.

OH X3C

O

OH

OH

63 X = H, F

j 10 Aldolases: Enzymes for Making and Breaking C----C Bonds

294

O

Aldolase DHAP

OR

H

O

OH

HO

OR

Reduction

OR

Deprotection

OR

P'ase

OH

OH

64

OH

O

HO

H OH 2-Deoxyaldose

O 2–

O

O3PS

OH

O

H

1. NaBH4

37 FruA

O

O

2. H3O+

O

=

O3PS

O

3. H2/Ni

OH

65

O HO

OH

O H

+

O OH

D-Olivose

4. HCl

OH

FSA

H3C HO HO

O

HO OH

OH 1-Deoxy-D-xylulose

66

Figure 10.25 Inverted approach for aldose synthesis using FruA catalysis, and application of the strategy for deoxysugar synthesis based on a phosphorothioate analog; synthesis of 1-deoxysugars by FSA catalyzed addition of hydroxyacetone.

maximize the yield of the preferred isomer (60) [102]. This general technique was applied in a novel approach for the de novo synthesis of 4,6-dideoxy sugars such as 4-deoxy-L-fucose or its triﬂuoromethylated analog (63); Figure 10.24) via stable ketose intermediates (62) [13]. On account of the structure of the DHAP nucleophile, the enzymatic aldolization technique is ideal for the direct synthesis of ketose monosaccharides and related derivatives or analogs. For an entry to aldoses, the inversion strategy has been developed (Figure 10.25), which utilizes monoprotected dialdehydes (e.g. (64)/(65)) for aldolization and, after stereoselective ketone reduction, provides free aldoses upon deprotection of the masked aldehyde function [129]. In this respect, the phosphorothioate analog (37) makes terminally deoxygenated sugars accessible via a sequence of FruA catalyzed aldolization followed by reductive desulfurization, as illustrated by the preparation of D-olivose along the inversion strategy [130]. Otherwise, deoxy sugars are usually only attained when the deoxy functionality is introduced with the aldehyde. An exception is the tolerance of FSA for hydroxyacetone as an aldol nucleophile, which gives rise to 1-deoxysugars such as (66) directly [91]. A more general access to biologically important and structurally more diverse aldose isomers makes use of ketol isomerases for the enzymatic interconversion of ketoses to aldoses. For a full realization of the concept of enzymatic stereodivergent carbohydrate synthesis, the stereochemically complementary L-rhamnose (RhaI; EC 5.3.1.14) and L-fucose isomerases (FucI; EC 5.3.1.3) from E. coli have been shown to display a relaxed substrate tolerance [16,99,113,131]. Both enzymes convert sugars and their derivatives that have a common (3R)-OH conﬁguration, but may deviate in

10.4 Dihydroxyacetone Phosphate Aldolases OH O

1. FucA, DHAP 2. P'ase

R

R HO

32

OH

O

FucI R

OH OH

O

OH HO OH

OH

j295

67

R = CH3, C2H5, CH=CH2, C≡CH OH O

32

1. RhuA , DHAP 2. P'ase

HO

OH

O

OH

RhaI

O

HO

OH

OH OH

OH

68

Figure 10.26 Short enzymatic synthesis of L-fucose and hydrophobic analogs, and of L-rhamnose, by aldolization–ketol isomerization, including kinetic resolution of racemic hydroxyaldehyde precursors.

stereochemistry or substitution pattern at subsequent positions of the chain [16,29]. As ketose products from RhuA and FucA catalyzed aldol reactions share the (3R) speciﬁcity, they can both be converted by the isomerases to corresponding aldose isomers, which provides for an access to a broad segment of aldose conﬁgurational space in a stereospeciﬁc, building block manner [29,132]. This strategy has been illustrated by tandem FucA–FucI catalysis in the synthesis of new L-fucose analogs (67) having tails with increased hydrophobicity and reactivity (Figure 10.26), starting from simple higher homologs of lactaldehyde and unsaturated analogs (32), as well as by the synthesis of L-rhamnose (68) and other L-conﬁgured aldohexoses using different enzyme combinations [113,131]. Similar results have been realized by utilizing a glucose isomerase (GlcI; EC 5.3.1.5), which is an industrially important enzyme for the isomerization of D-glucose to D-fructose. The latter enzyme has a more narrow speciﬁcity for D-fructose modiﬁcations but could be used in combined enzymatic syntheses, particularly of 6-modiﬁed D-glucose derivatives [133]. Several cyclitol derivatives of varying ring size, for example, (69)/(70), have been prepared based on an enzymatic aldolization as the initial step. Substrates carrying suitably installed C,H-acidic functional groups such as nitro, ester, phosphonate (or halogen) functionalities made use of facile intramolecular nucleophilic (or radical) cyclization reactions ensuing, or subsequent to, the enzyme-catalyzed aldol addition (Figure 10.27) [134–137]. Phosphonate analogs to phosphate esters, in which the PO bond is formally replaced by a PC bond, have attracted attention due to their stability toward the hydrolytic action of phosphatases, which renders them potential inhibitors or regulators of metabolic processes. Two alternative pathways, in fact, may achieve introduction of the phosphonate moiety by enzyme catalysis. The ﬁrst employs the bioisosteric methylene phosphonate analog (39), which yields products related to sugar 1-phosphates such as (71)/(72) (Figure 10.28) [102,107]. This strategy is rather effective because of the inherent stability of (39) as a replacement for (25), but depends on the individual tolerance of the aldolase for structural modiﬁcation close

j 10 Aldolases: Enzymes for Making and Breaking C----C Bonds

296

O 2N

AcO

1. FruA, DHAP 2. P'ase

OH

O 2N

O HO

CHO

AcO

OPO22–

CHO

OH

HO

2. P'ase

OH

OH

69

HO

NO2 NO2

NO2

AcO

Steps OH

HO

1. FruA, DHAP

OH

OH

NO2

HO

O

70

Figure 10.27 Preparation of aminocyclitol precursors by chemoenzymatic tandem reactions.

to the reactive center. The second option is the suitable choice of a phosphonylated aldehyde such as (73), which gives rise to analogs of sugar o-phosphates (74) [138,139]. The structural resemblance of azasugars (1-deoxy sugars in which an imino group replaces the ring oxygen) to transition states or intermediates of glyco-processing enzymes has made these compounds an attractive research object because of their potential value as enzyme inhibitors for therapeutic applications. An important and ﬂexible synthetic strategy has been developed which consists of a stereoselective enzymatic aldol addition to an azido aldehyde followed by azide hydrogenation with intramolecular reductive amination [140–142]. Particularly noteworthy are the stereodivergent chemoenzymatic syntheses of diastereomers of the nojirimycin type from

OH

O HO FruA O

O HO

+ H

RhuA

HO

O (EtO)2P

OH

FucA

( )n

1. FruA/TPI FBP

H O

2. P'ase

73 Figure 10.28 Complementary routes for the stereoselective synthesis of hydrolytically stable sugar phosphonates, either from the bioisosteric phosphonate analog of DHAP or from phosphonylated aldehydes.

PO32–

O OH OH

PO32–

39

71

HO HO

PO32–

O (EtO)2P

72

OH

( )n

O

OH

OH

HO

74 n = 1,2

10.4 Dihydroxyacetone Phosphate Aldolases OH NH

HO HO

HO 1-Deoxynojirimycin HO HO HN

OH OH

HO FruA

RhuA

HN

a

a

HO

RhuA

FruA

a

CHO

NH HO HO

HO HO HN

OH OH

FucA a

HO

D-75

HO HO

HO N3

N3

OH OH

OH OH NH

1-Deoxymannojirimycin

S

OH

OH

a

OHC

R

HO

j297

L-75

TagA a

TagA

FucA

a

a

OH

HN OH OH

HO

OH OH NH

HO

Figure 10.29 Stereodivergent synthesis of 1-deoxy azasugars of the nojirimycin type by two-step enzymatic aldolization/catalytic reductive amination (a¼ (i)DHAP, aldase; (ii)Pase; (iii)H2, Pd/C).

3-azidoglyceraldehyde (75) that have been developed independently by several groups (Figure 10.29) [85,117,143–146]. On account of the low kinetic selectivity of the rabbit FruA for 2-hydroxyaldehydes, use of enantiopure aldehyde proved superior to the racemate for preparation of the parent 1-deoxy-D-nojirimycin. An extensive array of further ﬁve-, six-, and seven-membered ring alkaloid analogs have since been made accordingly following the same general strategy. For structural variation, as exempliﬁed by (76)–(78), differently substituted azido aldehydes or Cbz-protected amino aldehydes of suitable chain length were converted by the distinct DHAP aldolases (Figure 10.30) [85,96,144,147–151]. Stereocontrol during the reductive cyclization seems to be effected best by Pd-catalyzed hydrogenation. Interestingly, the type and steric bulk of N-protecting groups may crucially inﬂuence the stereoselectivity of enzymatic aldol additions [151]. Most recently, a straightforward one-pot synthesis of D-fagomine and N-alkylated derivatives (80) has been realized by using FSA catalyzed addition of (31) to Cbz-protected aminopropanal (79), followed by reductive cyclization and ensuing N-alkylation [152]. The technique has been extended to the bifunctional class of azasugar phosphonic acids such as (81) by exploiting the tolerance of the rabbit FruA for the bioisosteric phosphonate nucleophile (39) [153]. In a strategy inverse to that employed for compound (81), FucA and FruA were employed in the chemoenzymic synthesis of six-membered iminocyclitol phosphonic acids [154]. Another illustrative example for the azasugar synthetic strategy concerns the chemoenzymatic synthesis of the naturally occurring australine, 3-epiaustraline, and 7-epialexin (Figure 10.31) [155]. The bicyclic pyrrolizidine core structure resulted from twofold reductive amination of a linear precursor (83) in which the asymmetric hydroxylation sites had been

j 10 Aldolases: Enzymes for Making and Breaking C----C Bonds

298

HO

H N S HO HO

1. FruA , DHAP

OH

HO

1. FruA, DHAP 2. P'ase 3. GlcI

OH O

2. P'ase 3. H2, Pd/C

N3

77

H2

O

HO HO

OH

HO

OH

OH

Pd/C

OH

N H

HO

CbzHN

OH

OH

31

O

RCHO

OH

CbzHN

H

FSA

N3

2–

PO3

OH

N3

O3P

OH

Pd/C

OH

80

H2 N+

H2

PO32–

FruA

H

OH

O

39

O

OH

OH

O 2–

N

H2, Pd/C

OH

79 HO

78

R

O O

OH

HO

N3

H

75

H HO S N HO HO

H

2. P'ase 3. H2, Pd/C

76

N3

1. FucA, DHAP

O S

HO

81

Figure 10.30 Stereoselective synthesis of five- and seven-membered ring azasugars, N-alkylated fagomine derivatives, and of novel azasugar phosphonates.

OH

1. FruA, DHAP 2. P'ase

O

OH

OH

OH

O

H OHCHN

O

OH OHCHN

82

OH

OHCHN

83

OH

Australine HO

HO

OH

1. O3 2. FruA, DHAP 3. P'ase

H N HO

HO

4. H2, Pd/C

OH

O HO

OH

HO

HO

85

Figure 10.31 Synthetic route to oxygenated pyrrolizidine alkaloids, and an aza-C-disaccharide as glycosidase inhibitors.

OH

N

3-Epiaustraline HO

OH

N3 (±) 84

HO OH

N

OH

H2, Pd/C

NaCNBH3

HO

OH

OH

OH

10.4 Dihydroxyacetone Phosphate Aldolases

installed during an aldolase-catalyzed chain extension from aminoaldehyde (82). A bidirectional aldolization approach furnished the C-glycosidically linked azadisaccharide (85) as an example of a disaccharide mimic. Ozonolysis of a racemic azidosubstituted cyclohexenediol precursor (84) was followed by tandem DHAP additions to both aldehydic termini to yield an intermediate azido-substituted dipyranoid 2,11-diulose which, when hydrogenated over Pd catalyst, highly selectively gave the aza-C-disaccharide (85) as a single diastereomer [13]. Such aldolase-catalyzed bidirectional chain elongation (tandem aldolization) of simple, readily available dialdehydes has been developed into an efﬁcient method for the generation of higher carbon sugars (e.g. (87)/(89)) by simple one-pot operations (Figure 10.32) [126,156]. The choice of furanoid (87) or pyranoid (89) nature of the products can be determined by a suitable hydroxyl substitution pattern in a corresponding cyclooleﬁnic precursor (86) versus (88)). The overall speciﬁc substitution

HO

OH

1. O3 2. FruA, DHAP 3. P'ase

HO

O R HO

HO HO

(±) 86

R

87

HO

OH

1. O3 2. FruA, DHAP 3. P'ase

OH

O HO

OH HO

OH O

HO HO

OH O

OH (±) 88

OH R

OH

R

HO

89 OH

OH OHC

(±)

HO

CHO

1. FruA, DHAP 2. P'ase

HO HO HO

OH

O O

OH OH

OH

1. O3 2. FruA, DHAP 3. P'ase

90

OH

Single diastereomer

O

O

HO HO HO

OH OH

91

OH OH

Figure 10.32 Applications of bidirectional chain extension for the synthesis of disaccharide mimetics and of annulated and spirocyclic oligosaccharide mimetics using tandem enzymatic aldol additions, including racemate resolution under thermodynamic control.

OH OH

j299

j 10 Aldolases: Enzymes for Making and Breaking C----C Bonds

300

O

OH H

O OPO32–

FruA

1. P'ase

O

OH O

OH

+ DHAP

O

2. H+

O

O

O O

92

O H

1. FruA, DHAP 2. P'ase

O

OH

HO OH

Acetone ZnI2

93

O HO O

O

Figure 10.33 Complementary, backbone-inverting approaches for the asymmetric synthesis of the insect pheromone (þ)-exobrevicomin.

pattern in the carbon-linked disaccharide mimetics is deliberately addressable by the relative hydroxyl conﬁguration and choice of the aldolase. Single diastereomers may be obtained in good overall yield from racemic precursors, if the tandem aldolizations are conducted under thermodynamic control (see Figure 10.24). Similarly, highly complex structures like annulated (90) and spirocyclic (91) carbohydrate mimics may be obtained from appropriately customized precursors (Figure 10.32) [126]. DHAP aldolases typically yield carbohydrates or carbohydrate-derived materials by nature of the reactive components, but they may also be advantageous in the construction of stereochemically homogeneous fragments of noncarbohydrate natural products. An impressive illustration is the FruA based chemoenzymatic syntheses of (þ)-exo-brevicomin (92); Figure 10.33), the aggregation pheromone of the Western pine bark beetle Dendroctonus brevicomis [157]. Addition of (25) to 5-oxohexanal generated an enantiopure vicinal syn-diol structure comprising the only independent stereogenic centers of brevicomin. A backbone-inverting approach toward (92) made use of 5,6-dideoxyketose precursor (93) [158], which is easily generated by FruA catalysis [84]. Application of an aldolase to the synthesis of the tricyclic microbial elicitor (–)syringolide (Figure 10.34) is another excellent example that enzyme-catalyzed aldolizations can be used to generate sufﬁcient quantities of enantiopure material in multistep syntheses of complex natural and unnatural products [159]. Remarkably, the aldolase reaction established absolute and relative conﬁguration of the only chiral centers that needed to be externally induced in the adduct (95) from achiral precursor (94); during the subsequent cyclization events, all others seemed to follow by kinetic preference.

10.4 Dihydroxyacetone Phosphate Aldolases O 1. DHAP, FruA H2O/DMF

O MPMO

H

2. P'ase

94

4 Steps

OH

O

O

OMPM

HO

O

CH3(CH2)6

O HO

j301

OH

O H+

95

55% CH3(CH2)6

OH

O

O

O O

(-)-Syringolide HO

Figure 10.34 Aldolase-based creation of two independent chiral centers in the total synthesis of the complex microbial plant defence elicitor ()-syringolide.

Using FruA catalysis and protected 4-hydroxybutanal, compound (97) has been stereoselectively prepared as a synthetic equivalent to the C-3–C-9 fragment of (þ)aspicillin, a lichen macrolactone (Figure 10.35) [160]. Similarly, FruA mediated stereoselective addition of (25) to a suitably crafted aldehyde precursor (98) served as the key step in the synthesis of the noncarbohydrate, skipped polyol C-9–C-16 chain fragment (99) of the macrolide antibiotic pentamycin [161,162]. A two-stage enzymatic sequence of arene dihydroxylation, using a naphthalene dioxygenase from Pseudomonas putida, followed by ozonolytic ring cleavage to yield dialdehyde (100) and RhuA-catalyzed aldolization has been developed for the synthesis of novel analogs of the cytotoxic pancratistatin pharmacophore such as (101) (Figure 10.36) [163]. This strategy converts a simple naphthalene core into

O H BnO

OH

1. FruA DHAP 2. P'ase

96

OH OH

BnO

O

OH

9 O

97

OH

O

3

OH

(+)-Aspicillin MeO

MeO

OBn

OBn

O

OMe H

98

1. FruA DHAP

11

HO

OH

OH

OH

OH

9

C5H11 HO

13

OMe

14 16

2. P'ase

HO

9

11

HO 15

O OH

99

13

HO

O

OH

14

O

16

CH3

HO 15 H3C

Figure 10.35 Stereoselective generation of chiral precursors for the synthesis of the lichen macrolactone (þ)-aspicillin and the macrolide antibiotic pentamycin using FruA catalysis.

Pentamycin

OH

j 10 Aldolases: Enzymes for Making and Breaking C----C Bonds

302

O O 1. NDO 2. O3 OH

OH CHO

O

OH CHO

O

RhuA

O

DHAP

O

OH

O OPO32–

OH CHO

OH

100 P'ase, Br2/CaCO3 OH HO O

OH NH

O OH

HO

OH

O

Pancratistatin

HO

OH

OH

OH O O

OH

+

O O

O

OH

O

OH

O O O

101

O

Figure 10.36 Enzyme-catalyzed asymmetric synthesis of a pancratistatin analog using a naphthalene dioxygenase and RhuA-catalyzed aldolization for the creation of four contiguous stereocenters.

a complex hybrid arene–carbohydrate structure, with simultaneous creation of four contiguous stereocenters, in just three steps.

10.5 Transketolase and Related Enzymes

Transketolase (EC 2.2.1.1) catalyzes the reversible transfer of a hydroxyacetyl nucleophile between various sugar phosphates using thiamine diphosphate as a cofactor. Enzymes from yeast, spinach, and E. coli have been shown to tolerate a broad spectrum of aldehyde acceptors, with the newly formed chiral center always possessing an absolute (S)-conﬁguration from re-face attack [164–166]. Although generic aldehydes are converted with full stereocontrol and even a,b-unsaturated aldehydes are acceptable to some extent, hydroxylated acceptors are usually converted at higher rates [167]. In addition, (2R)-antipodes of racemic 2-hydroxyaldehydes (32) are discriminated with complete enantioselectivity, which enables efﬁcient kinetic resolution (Figure 10.37) [168,169]. Vicinal D-threo diols (103) with (3S,4R)-conﬁguration are thus generated by a two-carbon chain elongation, which is stereochemically equivalent to the respective three-carbon elongation reaction achieved by FruA catalysis. For synthetic purposes, hydroxypyruvate (102) can effectively replace the natural sugar phosphate donors (albeit at reduced conversion rates [170]), which renders synthetic reactions irreversible owing to the loss of carbon dioxide [164–167].

10.5 Transketolase and Related Enzymes

H

R

OH

O

OH

+

OH

HO2C

Transketolase RR

O

-32

OH OH

+

H

R S

OH

CO2

102

D,L

O

S

j303

O

103

L

-32

Figure 10.37 Kinetic resolution by transketolase, and nonequilibrium CC bond formation by decomposition of hydroxypyruvate.

This thermodynamic driving force is particularly useful with multienzyme equilibrium systems such as that used in the gram-scale synthesis of two equivalents of D-xylulose 5-phosphate (104) from (26) (Figure 10.38) [171,172]. Similarly, the corresponding 1-deoxy-D-xylulose 5-phosphate was efﬁciently produced from pyruvate and (34) by the catalytic action of the thiamine diphosphate-dependent 1-deoxy-D-xylulose 5-phosphate synthase (DXS) (EC 2.2.1.7) from E. coli [173]. Transketolase has been utilized for the key steps in the chemoenzymatic syntheses of (þ)-exo-brevicomin (92) from racemic 2-hydroxybutyraldehyde via the intermediate (93) [158], as well as the azasugars 1,4-dideoxy-1,4-imino-D-arabinitol [117] or N-hydroxypyrrolidine (107) [174] from 3-azido (75) and 3-O-benzyl (105) derivatives of glyceraldehyde, respectively (Figure 10.39). Such syntheses were all conducted with intrinsic racemate resolution of 2-hydroxyaldehydes and proﬁted from the thermodynamic advantage of donor (102). For identiﬁcation of suitable mutant aldolases by engineering or evolutionary approaches, ﬂuorogenic compound (108) has been prepared as an efﬁcient screening substrate based on the high enantio- and diastereoselectivity of transketolase [175]. Further preparative applications of transketolase include the synthesis of valuable ketose sugars, particularly fructose analogs [164,165]. Other thiamine diphosphate-dependent enzymes have recently been scrutinized for their preparative value [166]. Although pyruvate decarboxylase (PDC) (EC 4.1.1.1) O 2– O

2– O

O HO

3PO

HO

OH

3PO

25

FruA OPO3=

OH TPI

26

OH 2–

TK

O3PO

18

OH 2–

O

CO2 –

O2 C

OH

104

O OH

102

Figure 10.38 Multienzymatic scheme for the stereoselective synthesis of two equivalents of xylulose 5-phosphate from fructose 1,6-bisphosphate.

O

O3PO OH

j 10 Aldolases: Enzymes for Making and Breaking C----C Bonds

304

OH H

102

O

N

107

Transketolase

OH

OH O

102

OH OH

OH

O

O

HO

Steps

OH

106

CO2

OH O

O

BnO

O

105

OH

Transketolase

BnO

O

CO2

O OH

O

108

OH

Figure 10.39 Synthesis of a novel N-hydroxypyrrolidine and a fluorogenic screening substrate for transaldolases based on stereospecific transketolase catalysis.

decarboxylates pyruvate to acetaldehyde in yeast fermentation, it can also catalyze the formation of (R)-phenylacetylcarbinol (PAC) (109) by ligating enzyme-bound acetaldehyde to externally added benzaldehyde (Figure 10.40) and its derivatives [176]. PAC is produced by this biotransformation on industrial scale as a key intermediate in the production of the pharmaceutical ephedrine and its diastereomer pseudoephedrine. Benzoylformate decarboxylase (BFD) (EC 4.1.1.7) catalyzes a stereoselective acylointype reaction upon decarboxylation of benzoylformate in the presence of acetaldehyde. Advantageously, the enzyme also accepts various aldehydes instead of the expensive ketoacid to yield to the corresponding (S)-acyloins, respectively, such as

O H

HO2C +

OH PDC R

O CO2

O +

109 O

H H

O

BFD

S

O

OH

110 O

O

H H

+

BAL

R

O

OH ent-110

Figure 10.40 Stereocomplementary acyloin syntheses based on the carboligation capacity of thiamine diphosphate-dependent enzymes.

10.6 2-Deoxy-D-Ribose 5-Phosphate Aldolase

O

OH 2–O

3PO

+

CHO

OH

RibA 2– O

H

H OH

O

O +

H

OH

RibA X

N3 D-75

O H

X

H3C

O

RibA

N3

112 O

X = H, F

OH

R

+

111

O

H3 C

H3C

O

3PO

18

H 3C

H HO

OH

CH3

113 O

O N3 D-75

H

N3

O

OH

H OH

RibA

HO

H2 Pd/C

j305

NH

HO HO

114

Figure 10.41 Natural aldol reaction catalyzed by RibA, acceptance of nonnatural aldol donors, and azasugar precursors prepared by stereoselective RibA catalysis.

(110) [177]. A biochemically related benzaldehyde lyase (BAL) (EC 4.1.2.38) catalyzes the same carboligation reactions, but with opposite (R)-selectivity (ent-110) [178]. All these enzymes seem to display a rather useful substrate tolerance for variously substituted aldehyde precursors.

10.6 2-Deoxy-D-Ribose 5-Phosphate Aldolase

The 2-deoxy-D-ribose 5-phosphate aldolase (RibA or DERA; EC 4.1.2.4) is a class I enzyme that in vivo catalyzes the reversible addition of ethanal to D-glyceraldehyde 3-phosphate (18); Figure 10.41) in the degradation of (111) from deoxyribonucleoside metabolism, with an equilibrium constant for synthesis of 4.2 · 103 M [34]. Hence, it is unique among the aldolases in that it uses an aldehyde rather than a ketone as the natural aldol donor. Interestingly, the E. coli enzymes relaxed acceptor speciﬁcity allows for substitution of both cosubstrates, albeit at strongly reduced (<1% of vmax) catalytic rates. Propanal, acetone, or ﬂuoroacetone can replace ethanal as the donor in the synthesis of variously substituted 3-hydroxyketones such as (112) or (113) (Figure 10.41)

j 10 Aldolases: Enzymes for Making and Breaking C----C Bonds

306

stage 1

DHAP FruA FBP

TPI O O 2–

H

O3PO

OH

H 2–

RibA

stage 2

O

O3PO

H OH

OH

18

111 N

Adenine

PPM

O

HO

OPO3H2

HO

Pi

N

N

PNP HO

NH2

O

HO

N

115

Figure 10.42 Two-stage aldolase-based process useful for the synthesis of deoxyribonucleosides.

[179,180]. Aldehydes up to a chain length of four nonhydrogen atoms are tolerated as acceptors. 2-Hydroxyaldehydes are relatively good acceptors, and the D-isomers are preferred over the L-isomers [180]. Reactions that lead to thermodynamically unfavorable structures may proceed with low stereoselectivity at the reaction center [181]. Recently, a single-point mutant aldolase was found 2.5 times more effective than the wild type in accepting unphosphorylated glyceraldehyde [182,183]. Inspired by its natural function, RibA has been applied in a multienzymatic commercial process for the production of different purine or pyrimidine containing deoxyribonucleosides such as (115) in good yield (Figure 10.42) [184,185]. Synthetic applications of RibA include the preparation of various sugar derivatives, such as 2-deoxy, thio and aza sugars [179,180]. Starting from azidoaldehydes, several azasugars containing a reduced hydroxylation density (e.g. (114); Figure 10.41) have been prepared by sequential aldolization–hydrogenation [180]. More recently, pyranose building blocks such as (117) have been prepared by using RibA catalysis that are useful as chiral key intermediates for the synthesis of anticancer agents epothilone A and C (Figure 10.43) [186]. When acetaldehyde is used as the only substrate, the initial aldol product can serve again as an acceptor for a sequential addition of a second donor molecule to give (3R,5R)-2,4,6-trideoxyhexose (118) (Figure 10.44) [187,188]. Cyclization to a stable hemiacetal masks the free aldehyde and thus effectively precludes formation of higher order adducts. When the ﬁrst acceptor is an a-substituted acetaldehyde, related aldol products from twofold donor additions can be prepared that are structurally related to mevinic acid lactone. The reaction with chloroethanal, which is a particular good acceptor substrate, leading to (119) has most recently been developed into an industrial process for the large-scale manufacturing of the skipped polyol, chiral statin side chain in cholesterol-lowering drugs related to compactin (Figure 10.44) [189–191]. Combination of two aldolases for sequential aldol additions, such as RibA for the initial and FruA or NeuA for a consecutive reaction, has also been studied for the synthesis of unnatural sugars [181,187].

10.6 2-Deoxy-D-Ribose 5-Phosphate Aldolase O

O

RibA

+

HO

116

O

O

OH

O

OH

117

OH

O

S

HO

N PMPO

O O

OH

O

OtBu

Epothilone A

O

OTBS O

Figure 10.43 Chemoenzymatic synthesis of a chiral eopthilone A building block based on an aldolization reaction.

O

+

j307

O

O

+

RibA

OH

OH

R

R

O

OH

OH

O

H

Cl S

RibA

HO

O

HO

O

Oxidation

OH

S

R

HO

Compactin

Figure 10.44 RibA catalyzed sequential aldol additions yielding a key chiral building block for cholesterol-lowering drugs.

O

OH

O Cl

O

OH Cl

RibA

O

O

118

O H

Cl

OH

H

O O

O

O Cl

119

j 10 Aldolases: Enzymes for Making and Breaking C----C Bonds

308

NH2

O

SHMT

+

H

CO2H

H

NH2 CO2H

S

PLP

120

OH

121 NH2

O

NH2 +

H3C

H

CO2H

120

ThrA

H3C R

S

CO2H

NH2 +

H3C S

OH

PLP

122

L-Thr

S

CO2H

OH

123

L- allo-Thr

Figure 10.45 Aldol reactions catalyzed in vivo by serine hydroxymethyl transferase and by threonine aldolases.

10.7 Glycine Aldolases

The metabolism of b-hydroxy-a-amino acids involves pyridoxal phosphate–dependent enzymes, classiﬁed as serine hydroxymethyltransferase (SHMT) (EC 2.1.2.1) or threonine aldolases (ThrA; L-threonine selective ¼ EC 4.1.2.5, L-allo-threonine selective ¼ EC 4.1.2.6). Both enzymes catalyze reversible aldol-type cleavage reactions yielding glycine (120) and an aldehyde (Figure 10.45) [192]. Whereas SHMT in vivo has a biosynthetic function, threonine aldolase catalyzes the degradation of threonine; both L- and D-speciﬁc ThrA enzymes are known [16,192]. Typically, ThrA enzymes show complete enantiopreference for their natural a-D- or a-L-amino conﬁguration but, with few exceptions, have only low speciﬁcity for the relative threo/erythro-conﬁguration (e.g. (122)/(123)) [193]. Likewise, SHMT is highly selective for the L-conﬁguration, but has poor threo/erythro selectivity [194]. For biocatalytic applications, the known SHMT, D- and L-ThrA show broad substrate tolerance for various acceptor aldehydes, notably including aromatic aldehydes [193–196]; however, a,b-unsaturated aldehydes are not accepted [197]. For preparative reactions, excess of (120) must compensate for the unfavorable equilibrium constant [34] to achieve economical yields. b-Hydroxyamino acids constitute an important class of compounds. The synthesis of L-serine (121) has been developed at multimolar scale using SHMT from Klebsiella aerogenes or E. coli to furnish the product at high ﬁnal concentration of >450 g L1 [198,199]. SHMT has also been employed for the synthesis of L-erythro-2-amino-3hydroxy-1,6-hexanedicarboxylic acid (131) [200], a potential precursor for carbocyclic b-lactams and nucleotides (Figure 10.46). The erythro-selective L-ThrA from the yeast Candida humicola has been used for the synthesis of (S,S,R)- and (S,S,S)-3,4dihydroxyprolines (128)/(129) from D-glyceraldehyde acetonide [201]. The same enzyme has also been applied to the synthesis of nucleobase-modiﬁed amino acids of type (127) [202] and for the preparation of a chiral building block (130) toward the synthesis of the immunosuppressive lipid mycestericin D [203,204]. Under kinetic control, the synthesis of novel sialyl Lewisx mimics could be conducted stereoselectively

10.7 Glycine Aldolases OH CHO

ThrA

O O

OH CO2H

O O

Glycine

+

NH2

124

CO2H

O O

NH2

125

126 Steps

Steps OH HO

CO2H N N

NH2

N N

N H

127

HO

OH CO2H

OH CO2H

N H

128

129

NH2

(a)

O H

OH

L-ThrA

OBn

OH OBn

H 2N

Glycine

CO2H

95

H2N

CO2H CO2H

130

131

OH H2N HOH2C

C6H13 CO2H

O

D-ThrA

H H3CO2S

(b)

Mycestericin D

O

OH

OH CO2H

Glycine

NH2

H3CO2S

H3CO2S

132

OH HN

CHCl2

Thiamphenicol O

Figure 10.46 Application of ThrA catalysis for the stereoselective synthesis of dihydroxyprolines from glyceraldehyde, and an adenylamino acid for RNA mimics (a). ThrA based preparation of precursors to the immunosuppressive lipid mycestericin and the antibiotic thiamphenicol (b).

[205]. Phenylserine derivative (132), precursor to the enantiomer of the antibiotic thiamphenicol, has been prepared with 92% de and >99% ee using a recombinant D-ThrA from Alcaligenes xylosoxidans, whereas the opposite L-conﬁgurated isomer was obtained by L-ThrA catalysis with only low diastereoselectivity [196]. Owing to the fully reversible equilibrium nature of the aldol addition process, enzymes with low diastereoselectivity will typically lead to a thermodynamically controlled mixture of erythro/threo-isomers that are difﬁcult to separate. The thermodynamic origin of poor threo/erythro selectivity has most recently been turned to an asset by the design of a diastereoselective dynamic kinetic resolution process by coupling of L-ThrA and a diastereoselective L-tyrosine decarboxylase (Figure 10.47)

j309

j 10 Aldolases: Enzymes for Making and Breaking C----C Bonds

310

OH

OH CO2H

L-ThrA

L-TyrD

NH2

CHO

NH2

Fast CO2

L- threo-133

(R)-134

Fast

+ 120

OH

OH CO2H

L-ThrA

L-TyrD

NH2

NH2

Slow CO2

133

L-erythro-

(S)-134

(a)

OH

OH CO2H

NH2

X DL-threo

R

L-ThrA

S

D-threo

(b)

NH2 DL-threo-

136

120

OH CO2H

O

+

X

-135

OH O

+

NH2

X

-135

CHO

CO2H

D-ThrA

CO2H

O

NH2

O L-threo

+

Piperonal +

120

-136

Figure 10.47 Dynamic kinetic resolution of ThrA generated diastereomers by enantioselective decarboxylation (a). Conventional kinetic resolution of diastereomer mixtures by retroaldolization for preparation of enantiopure arylserines and for a synthetic intermediate of an antiparkinsonism drug (b).

[206]. By this concept, a reversible enzymatic aldol reaction generates a mixture of Lthreo/erythro aldol diastereomers (133) from which the L-threo isomer is preferentially decomposed by an irreversible decarboxylation to furnish aromatic aminoalcohol (R)(134) with 78% ee in high yield. The speciﬁcity of SHMT or ThrA enzymes can be exploited better for kinetic resolutions of stereoisomer mixtures such as those produced by chemical synthesis (Figure 10.47). This is particularly promising for aryl analogs of threonine that are of interest as building blocks of pharmaceuticals, including vancomycin antibiotics. Thus, an L-ThrA from Streptomyces amakusaensis has been shown to be particularly useful for the resolution of racemic threo-aryl serines (135) by retroaldolization under kinetic control to furnish enantiomerically pure D-threo-amino acids [193,200,207,208]. As a complementary example, the recombinant low-speciﬁcity D-ThrA from A. xylosoxidans has been used for the resolution of DL-threo-b-(3,4methylenedioxyphenyl)serine (136) by retro-aldol cleavage to furnish the desired L-threo isomer with a molar yield of 50% and almost 100% ee [209]. The latter

References

compound serves as a synthetic intermediate en route to L-threo-p-(3,4-dihydroxyphenyl)serine (DOPS), a parkinsonism drug.

10.8 Summary

The aldol addition is one of the most fundamental methodologies available to the synthetic chemist for the construction of new carbon–carbon bonds. Progress in the preparative use of aldol active enzymes as efﬁcient and highly stereoselective catalysts has been remarkable. Biocatalytic CC bond ligation is indeed eminently useful, and highly predictable, for the asymmetric synthesis of complex multifunctional molecules. This chapter has attempted to demonstrate the potential of enzymatic aldol reactions by highlighting the state of development of readily available enzymes, the most important synthetic examples, and the most efﬁcient reaction techniques. It is also evident that the technology is well accepted in the chemical community and, with several examples for large-scale industrial processes now in operation, that the ﬁeld is maturing rapidly. The scope of potential applications is broadening on all frontiers with increasing complexity of target structures. However, it becomes obvious that enzymatic aldol reactions still suffer from restrictions that derive from a limited ﬂexibility of enzymes for their donor components. In order to remain competitive with modern chemical developments such as proline-based organocatalysis [210], opportunities to lift such restraints by tailoring aldolases for speciﬁc needs by rational protein engineering based on a host of available protein crystal structures [16] or by directed evolution via random mutagenesis [211] are therefore currently in focus. Prospecting for novel activities from whole-genome sequencing efforts or by mass screening of previously untapped biodiversity may likewise furnish attractive biocatalysts for new applications [212,213].

References 1 Wong, C.-H. and Whitesides, G.M. (1994) Enzymes in Synthetic Organic Chemistry, Pergamon, Oxford. 2 Sinnott, M. (1998) Comprehensive Biological Catalysis, Academic Press, San Diego. 3 Fessner, W.-D. (1998) Biocatalysis - From Discovery to Application Topics in Current Chemistry, Vol. 200, Springer, Heidelberg. 4 Patel, R.N. (2000) Stereoselective Biocatalysis, Marcel Dekker Inc, New York. 5 Drauz, K. and Waldmann, H. (2002) Enzyme Catalysis in Organic Synthesis, 2nd edn, Wiley-VCH, Weinheim.

6 Faber, K. (2004)Biotransformations in Organic Chemistry, 5th edn, Springer, Heidelberg. 7 Bommarius, A.S. and Riebel, B.R. (2004) Biocatalysis, Wiley-VCH, Weinheim. 8 Liese, A., Seelbach, K. and Wandrey, C. (2000)Industrial Biotransformations, Wiley-VCH, Weinheim. 9 Aehle, W. (2003) Enzymes in Industry: Production and Applications, Wiley-VCH, Weinheim. 10 Patel, R.N. (2007) Biocatalysis in the Pharmaceutical and Biotechnology Industries, CRC Press, Boca Raton.

j311

j 10 Aldolases: Enzymes for Making and Breaking C----C Bonds

312

11 Schmid, A., Dordick, J.S., Hauer, B., Kiener, A., Wubbolts, M. and Withold, B. (2001) Nature, 409, 258–268. 12 Schoemaker, H.E.,Mink, D.and Wubbolts, M.G. (2003) Science, 299, 1694–1697. 13 Fessner, W.-D. and Walter, C. (1996) Topics in Current Chemistry, 184, 97–194. 14 Fessner, W.-D. (1998) Current Opinion in Chemical Biology, 2, 85–97. 15 Fessner, W.-D. and Helaine, V. (2001) Current Opinion in Biotechnology, 12, 574–586. 16 Fessner, W.-D. (2004) In Modern Aldol Reactions, (ed. R. Mahrwald), Vol. 1, Wiley-VCH, Weinheim. pp. 201–272. 17 Silvestri M.G., DeSantis, G., Mitchell, M. and Wong, C.-H. (2003) Topics in Stereochemistry, 23, 267–342. 18 Sukumaran, J. and Hanefeld, U. (2005) Chemical Society Reviews, 34, 530–542. 19 Samland, A.K. and Sprenger, G.A. (2006) Applied Microbiology and Biotechnology, 71, 253–264. 20 Machajewski, T.D. and Wong, C.-H. (2000) Angewandte Chemie International Edition, 39, 1352–1374. 21 Mahrwald, R. (2004) Modern Aldol Reactions, Wiley-VCH Verlag GmbH, Weinheim. 22 Palomo, C., Oiarbide, M. and Garcia, J.M. (2004) Chemical Society Reviews, 33, 65–75. 23 Schomburg, D. and Salzmann, M. (1990–1995) Enzyme Handbook, Springer, Berlin. 24 Henderson, D.F. and Toone, E.J. (1999)Comprehensive Natural Product Chemistry, (ed. B.M. Pinto), Vol. 3, Elsevier Science, Amsterdam pp. 367–440. 25 Toone, E.J., Simon, E.S., Bednarski, M.D. and Whitesides, G.M. (1989) Tetrahedron, 45, 5365–5422. 26 David, S., Auge, C. and Gautheron, C. (1991) Advances in Carbohydrate Chemistry and Biochemistry, 49, 175–237. 27 Gijsen, H.J.M., Qiao, L., Fitz, W. and Wong, C.-H. (1996) Chemical Reviews, 96, 443–473.

28 Wymer, N. and Toone, E.J. (2000) Current Opinion in Chemical Biology, 4, 110–119. 29 Fessner, W.-D. (1992) Microbial Reagents in Organic Synthesis (ed. S. Servi), Vol. 381, Kluwer Academic Publishers, Dordrecht, pp. 43–55. 30 Horecker, B.L., Tsolas, O. and Lai, C.Y. (1972) The Enzymes, (ed. P.D. Boyer), Vol. VII, Academic Press, New York, pp. 213–258. 31 Kluger, R. (1990) Chemical Reviews, 90, 1151–1169. 32 Kuo, D.J. and Rose, I.A. (1985) Biochemistry, 24, 3947–3952. 33 Fessner, W.-D., Schneider, A., Held, H., Sinerius, G., Walter, C., Hixon, M., and Schloss, J.V. (1996) Angewandte ChemieInternational Edition in English, 35, 2219–2221. 34 Goldberg, R.N. and Tewari, Y.B. (1995) Journal of Physical and Chemical Reference Data, 24, 1669–1698. 35 Fessner, W.-D. and Knorst, M. (2002) DE 10034586; Chemical Abstracts, 136, 166160. 36 Kim,M.J.,Hennen,W.J.,Sweers,H.M.and Wong, C.-H. (1988) Journal of the American Chemical Society, 110, 6481–6486. 37 Sugai, T., Kuboki, A., Hiramatsu, S., Okazaki, H. and Ohta, H. (1995) Bulletin of the Chemical Society of Japan, 68, 3581–3589. 38 Blayer, S., Woodley, J.M., Lilly, M.D. and Dawson, M.J. (1996) Biotechnology Progress, 12, 758–763. 39 Mahmoudian, M., Noble, D., Drake, C.S., Middleton, R.F., Montgomery, D.S., Piercey, J.E., Ramlakhan, D., Todd, M. and author>Dawson, M.J. (1997) Enzyme and Microbial Technology, 20, 393–400. 40 Blayer, S., Woodley, J.M., Dawson, M.J. and Lilly, M.D. (1999) Biotechnology and Bioengineering, 66, 131–136. 41 Kragl, U., Gygax, D., Ghisalba, O. and Wandrey, C. (1991) Angewandte ChemieInternational Edition in English, 30, 827–828.

References 42 Maru, I., Ohnishi, J., Ohta, Y. and Tsukada, Y. (1998) Carbohydrate Research, 306, 575–578. 43 Lee, J.-O., Yi, J.-K., Lee, S.-G., Takahashi, S. and Kim, B.G. (2004) Enzyme and Microbial Technology, 35, 121–125. 44 Woodhall, T., Williams, G., Berry, A. and Nelson, A. (2005) Angewandte Chemie International Edition, 44, 2109–2112. 45 Williams, G.J., Woodhall, T., Farnsworth, L.M., Nelson, A. and Berry, A. (2006) Journal of the American Chemical Society, 128, 16238–16247. 46 Tabata, K., Koizumi, S., Endo, T. and Ozaki, A. (2002) Enzyme and Microbial Technology, 30, 327–333. 47 Salagnad, C., Godde, A., Ernst, B. and Kragl, U. (1997) Biotechnology Progress, 13, 810–813. 48 Kragl,U.,G€ odde,A.,Wandrey,C.,Lubin,N. and Auge, C. (1994) Journal of the Chemical Society, Perkin Transactions1, 119–124. 49 Beliczey, J., Kragl, U., Liese, A., Wandrey, C., Hamacher, K., Coenen, H.H. and Tierling, T. (2002) US 6355453; Chemical Abstracts, 136, 231340. 50 Auge, C., David, S., Gautheron, C., Malleron, A. and Cavaye, B. (1988) New Journal of Chemistry, 12, 733–744. 51 Fitz, W., Schwark, J.R. and Wong, C.-H. (1995) Journal of Organic Chemistry, 60, 3663–3670. 52 Auge, C., Bouxom, B., Cavaye, B. and Gautheron, C. (1989) Tetrahedron Letters, 30, 2217–2220. 53 Liu, J.L.C., Shen, G.J., Ichikawa, Y., Rutan, J.F., Zapata, G., Vann, W.F. and Wong, C.-H. (1992) Journal of the American Chemical Society, 114, 3901–3910. 54 Lins, R.J., Flitsch, S.L., Turner, N.J., Irving, E. and Brown, S.A. (2002) Angewandte Chemie International Edition, 41, 3405–3407. 55 Lins, R.J., Flitsch, S.L., Turner, N.J., Irving, E. and Brown, S.A. (2004) Tetrahedron, 60, 771–780. 56 Wu, W.-Y., Jin, B., Kong, D.C.M. and von Itzstein, M. (1997) Carbohydrate Research, 300, 171–174.

57 Kuboki,A., Okazaki,H., Sugai, T. and Ohta, H. (1997) Tetrahedron, 53, 2387–2400. 58 Humphrey, A.J., Fremann, C., Critchley, P., Malykh, Y., Schauer, R. and Bugg, T.D.H. (2002) Bioorganic and Medicinal Chemistry, 10, 3175–3185. 59 Pan, Y., Ayani, T., Nadas, J., Wen, S. and Guo, Z. (2004) Carbohydrate Research, 339, 2091–2100. 60 Lin, C.-C., Lin, C.-H. and Wong, C.-H. (1997) Tetrahedron Letters, 38, 2649–2652. 61 Murakami, M., Ikeda, K. and Achiwa, K. (1996) Carbohydrate Research, 280, 101–110. 62 Kong, D.C.M. and von Itzstein, M. (1997) Carbohydrate Research, 305, 323–329. 63 Kiefel, M.J., Wilson, J.C., Bennett, S., Gredley, M. and von Itzstein, M. (2000) Bioorganic and Medicinal Chemistry, 8, 657–664. 64 Sparks, M.A., Williams, K.W., Lukacs, C., Schrell, A., Priebe, G., Spaltenstein, A. and Whitesides, G.M. (1993) Tetrahedron, 49, 1–12. 65 Zhou, P.Z., Salleh, H.M. and Honek, J.F. (1993) Journal of Organic Chemistry, 58, 264–266. 66 Fitz, W. and Wong, C.-H. (1994) Journal of Organic Chemistry, 59, 8279–8280. 67 Ikeda, K., Mori, H. and Sato, M. (2006) Chemical Communications, 29, 3093–3094. 68 Lin, C.H., Sugai, T., Halcomb, R.L., Ichikawa, Y. and Wong, C.-H. (1992) Journal of the American Chemical Society, 114, 10138–10145. 69 David, S., Malleron, A. and Cavaye, B. (1992) New Journal of Chemistry, 16, 751–755. 70 Auge, C., Gautheron, C., David, S., Malleron, A., Cavaye, B. and Bouxom, B. (1990) Tetrahedron, 46, 201–214. 71 Sugai, T., Shen, G.J., Ichikawa, Y. and Wong, C.-H. (1993) Journal of the American Chemical Society, 115, 413–421. 72 Koppert, K. and Brossmer, R. (1992) Tetrahedron Letters, 33, 8031–8034. 73 Malleron, A. and David, S. (1996) New Journal of Chemistry, 20, 153–159.

j313

j 10 Aldolases: Enzymes for Making and Breaking C----C Bonds

314

74 Ghalambor, M.A. and Heath, E.C. (1966) Journal of Biological Chemistry, 241, 3222–3227. 75 Wood, W.A. (1972) The Enzymes, (ed. P.D. Boyer), 3rd edn, Vol. 7, Academic Press, New York, pp. 281–302. 76 Floyd, N.C., Liebster, M.H. and Turner, N.J. (1992) Journal of the Chemical Society, Perkin Transactions 1, 1085–1086. 77 Allen, S.T., Heintzelman, G.R. and Toone, E.J. (1992) Journal of Organic Chemistry, 57, 426–427. 78 Shelton, M.C., Cotterill, I.C., Novak, S.T.A., Poonawala, R.M., Sudarshan, S. and Toone, E.J. (1996) Journal of the American Chemical Society, 118, 2117–2125. 79 Cotterill, I.C., Shelton, M.C., Machemer, D.E.W., Henderson, D.P. and Toone, E.J. (1998) Journal of the Chemical Society, Perkin Transactions 1, 1335–1341. 80 Fong, S., Machajewski, T.D., Mak, C.C. and Wong, C.-H. (2000) Chemistry & Biology, 7, 873–883. 81 Wymer, N., Buchanan, L.V., Henderson, D., Mehta, N., Botting, C.H., Pocivavsek, L., Fierke, C.A., Toone, E.J. and Naismith, J.H. (2001) Structure, 9, 1–10. 82 Henderson, D.P., Shelton, M.C., Cotterill, I.C. and Toone, E.J. (1997) Journal of Organic Chemistry, 62, 7910–7911. 83 Henderson, D.P., Cotterill, I.C., Shelton, M.C. and Toone, E.J. (1998) Journal of Organic Chemistry, 63, 906–907. 84 Bednarski, M.D., Simon, E.S., Bischofberger, N., Fessner, W.-D., Kim, M.J., Lees, W., Saito, T., Waldmann, H. and Whitesides, G.M. (1989) Journal of the American Chemical Society, 111, 627–635. 85 von der Osten, C.H., Sinskey, A.J., Barbas, C.F., Pederson, R.L., Wang, Y.F. and Wong, C.-H. (1989) Journal of the American Chemical Society, 111, 3924–3927. 86 Fessner, W.-D., Sinerius, G., Schneider, A., Dreyer, M., Schulz, G.E., Badia, J. and Aguilar, J. (1991) Angewandte Chemie-International Edition in English, 30, 555–558.

87 Zannetti, M.T., Walter, C., Knorst, M. and Fessner, W.-D. (1999) Chemistry- A European Journal, 5, 1882–1890. 88 Espelt, L., Clapes, P., Esquena, J., Manich, A. and Solans, C. (2003) Langmuir, 19, 1337–1346. 89 Lees, W.J. and Whitesides, G.M. (1993) Journal of Organic Chemistry, 58, 1887–1894. 90 Sch€ urmann, M. and Sprenger, G.A. (2001) Journal of Biological Chemistry, 276, 11055–11061. 91 Sch€ urmann, M., Sch€ urmann, M. and Sprenger, G.A. (2002) Journal of Molecular Catalysis B: Enzymatic, 19–20, 247–252. 92 Bissett, D.L. and Anderson, R.L. (1980) Journal of Biological Chemistry, 255, 8750–8755. 93 Fessner, W.-D. and Eyrisch, O. (1992) Angewandte Chemie-International Edition in English, 31, 56–58. 94 Eyrisch, O., Sinerius, G. and Fessner, W.-D. (1993) Carbohydrate Research, 238, 287–306. 95 Williams, G.J., Domann, S., Nelson, A. and Berry, A. (2003) Proceedings of the National Academy of Sciences of the United States of America, 100, 3143–3148. 96 Espelt, L., Parella, T., Bujons, J., Solans, C., Joglar, J., Delgado, A. and Clapes, P. (2003) Chemistry- A European Journal, 9, 4887–4899. 97 Ozaki, A., Toone, E.J., von der Osten, C.H., Sinskey, A.J. and Whitesides, G.M. (1990) Journal of the American Chemical Society, 112, 4970–4971. 98 Espelt, L., Bujons, J., Parella, T., Calveras, J., Joglar, J., Delgado, A. and Clapes, P. (2005) Chemistry- A European Journal, 11, 1392–1401. 99 Fessner, W.-D., Badia, J., Eyrisch, O., Schneider, A. and Sinerius, G. (1992) Tetrahedron Letters, 33, 5231–5234. 100 Fessner, W.-D., Schneider, A., Eyrisch, O., Sinerius, G. and Badia, J. (1993) Tetrahedron: Asymmetry, 4, 1183–1192.

References 101 Fessner, W.-D. and Walter, C. (1992) Angewandte Chemie-International Edition in English, 31, 614–616. 102 Fessner, W.-D. and Sinerius, G. (1994) Angewandte Chemie-International Edition in English, 33, 209–212. 103 Hettwer, J., Oldenburg, H. and Flaschel, E. (2002) Journal of Molecular Catalysis B: Enzymatic, 19–20, 215–222. 104 Schoevaart, R., van Rantwijk, F. and Sheldon, R.A. (2000) Journal of Organic Chemistry, 65, 6940–6943. 105 Charmantray, F., Dellis, P., Samreth, S. and Hequet, L. (2006) Tetrahedron Letters, 47, 3261–3263. 106 Arth, H.L., Sinerius, G. and Fessner, W.-D. (1995) Liebigs Annalen, 2037–2042. 107 Arth, H.L. and Fessner, W.-D. (1997) Carbohydrate Research, 305, 313–321. 108 Drueckhammer, D.G., Durrwachter, J.R., Pederson, R.L., Crans, D.C., Daniels, L. and Wong, C.-H. (1989) Journal of Organic Chemistry, 54, 70–77. 109 Schoevaart, R., van Rantwijk, F. and Sheldon, R.A. (2001) Journal of Organic Chemistry, 66, 4559–4562. 110 Bednarski, M.D., Waldmann, H.J. and Whitesides, G.M. (1986) Tetrahedron Letters, 27, 5807–5810. 111 Franke, D., Machajewski, T., Hsu, C.-C. and Wong, C.-H. (2003) Journal of Organic Chemistry, 68, 6828–6831. 112 Kim, M.-J. and Lim, I.T. (1996) Synlett, 138–140. 113 Fessner, W.-D., Gosse, C., Jaeschke, G. and Eyrisch, O. (2000) European Journal of Organic Chemistry, 125–132. 114 Wong, C.-H. and Whitesides, G.M. (1983) Journal of Organic Chemistry, 48, 3199–3205. 115 Maliakel, B.P. and Schmid, W. (1993) Journal of Carbohydrate Chemistry, 12, 415–424. 116 Azema, L., Bringaud, F., Blonski, C. and Perie, J. (2000) Bioorganic and Medicinal Chemistry, 8, 717–722. 117 Ziegler, T., Straub, A. and Effenberger, F. (1988) Angewandte Chemie-International Edition in English, 27, 716–717.

118 Nicotra, F., Panza, L., Russo, G. and Verani, A. (1993) Tetrahedron: Asymmetry, 4, 1203–1204. 119 Zhu, W. and Li, Z. (2000) Perkin Transactions 1, 1105–1108. 120 Liu, K.K.C. and Wong, C.-H. (1992) Journal of Organic Chemistry, 57, 4789–4791. 121 Turner, N.J. and Whitesides, G.M. (1989) Journal of the American Chemical Society, 111, 624–627. 122 Crestia, D., Guerard, C., Bolte, J. and Demuynck, C. (2001) Journal of Molecular Catalysis B: Enzymatic, 11, 207–212. 123 Sevestre, A., Helaine, V., Guyot, G., Martin, C. and Hecquet, L. (2003) Tetrahedron Letters, 44, 827–830. 124 Wong, C.-H., Moris-Varas, F., Hung, S.-C., Marron, T.G., Lin, C.-C., Gong, K.W. and Weitz-Schmidt, G. (1997) Journal of the American Chemical Society, 119, 8152–8158. 125 Eyrisch, O., Keller, M. and Fessner, W.-D. (1994) Tetrahedron Letters, 35, 9013–9016. 126 Petersen, M., Zannetti, M.T. and Fessner, W.-D. (1997) Topics in Current Chemistry, 186, 87–117. 127 Durrwachter, J.R. and Wong, C.-H. (1988) Journal of Organic Chemistry, 53, 4175–4181. 128 Liu, K.K.C., Pederson, R.L. and Wong, C.-H. (1991) Journal of the Chemical Society. Perkin Transactions 1, 2669–2673. 129 Borysenko, C.W., Spaltenstein, A., Straub, J.A. and Whitesides, G.M. (1989) Journal of the American Chemical Society, 111, 9275–9276. 130 Duncan, R. and Drueckhammer, D.G. (1996) Journal of Organic Chemistry, 61, 438–439. 131 Wong, C.-H., Alajarin, R., Moris-Varas, F., Blanco, O. and Garcia-Junceda, E. (1995) Journal of Organic Chemistry, 60, 7360–7363. 132 Fessner, W.-D. (1993) GIT Fachzeitschrift fuer das Laboratorium, 37, 951–956.

j315

j 10 Aldolases: Enzymes for Making and Breaking C----C Bonds

316

133 Durrwachter, J.R., Drueckhammer, D.G., Nozaki, K., Sweers, H.M. and Wong, C.-H. (1986) Journal of the American Chemical Society, 108, 7812–7818. 134 Schmid, W. and Whitesides, G.M. (1990) Journal of the American Chemical Society, 112, 9670–9671. 135 Chou, W.-C., Fotsch, C. and Wong, C.-H. (1995) Journal of Organic Chemistry, 60, 2916–2917. 136 Gijsen, H.J.M. and Wong, C.-H. (1995) Tetrahedron Letters, 36, 7057–7060. 137 El Blidi, L., Crestia, D., Gallienne, E., Demuynck, C., Bolte, J. and Lemaire, M. (2004) Tetrahedron: Asymmetry, 15, 2951–2954. 138 Guanti, G., Banﬁ, L. and Zannetti, M.T. (2000) Tetrahedron Letters, 41, 3181–3185. 139 Guanti, G., Zannetti, M.T., Banﬁ, L. and Riva, R. (2001) Advanced Synthesis and Catalysis, 343, 682–691. 140 Look, G.C., Fotsch, C.H. and Wong, C.-H. (1993) Accounts of Chemical Research, 26, 182–190. 141 Wong, C.-H., Halcomb, R.L., Ichikawa, Y. and Kajimoto, T. (1995) Angewandte Chemie-International Edition in English, 34, 412–432. 142 Whalen, L.J. and Wong, C.-H. (2006) Aldrichimica Acta, 39, 63–71. 143 Pederson, R.L., Kim, M.-J. and Wong, C.-H. (1988) Tetrahedron Letters, 29, 4645–4648. 144 Liu, K.K.C., Kajimoto, T., Chen, L., Zhong, Z., Ichikawa, Y. and Wong, C.-H. (1991) Journal of Organic Chemistry, 56, 6280–6289. 145 Zhou, P.Z., Salleh, H.M., Chan, P.C.M., Lajoie, G., Honek, J.F., Nambiar, P.T.C. and Ward, O.P. (1993) Carbohydrate Research, 239, 155–166. 146 Lees, W.J. and Whitesides, G.M. (1992) Bioorganic Chemistry, 20, 173–179. 147 Straub, A., Effenberger, F. and Fischer, P. (1990) Journal of Organic Chemistry, 55, 3926–3932. 148 Hung, R.R., Straub, J.A. and Whitesides, G.M. (1991) Journal of Organic Chemistry, 56, 3849–3855.

149 Moris-Varas, F., Qian, X.-H. and Wong, C.-H. (1996) Journal of the American Chemical Society, 118, 7647–7652. 150 Mitchell, M.L., Lee, L.V. and Wong, C.-H. (2002) Tetrahedron Letters, 43, 5691–5693. 151 Calveras, J., Bujons, J., Parella, T., Crehuet, R., Espelt, L., Joglar, J. and Clapes, P. (2006) Tetrahedron, 62, 2648–2656. 152 Castillo, J.A., Calveras, J., Casas, J., Mitjans, M., Vinardell, M.P., Parella, T., Inoue, T., Sprenger, G.A., Joglar, J. and Clapes, P. (2006) Organic Letters, 8, 6067–6070. 153 Schuster, M., He, W.F. and Blechert, S. (2001) Tetrahedron Letters, 42, 2289–2291. 154 Mitchell, M., Qaio, L. and Wong, C.-H. (2001) Advanced Synthesis and Catalysis, 343, 596–599. 155 Romero, A. and Wong, C.-H. (2000) Journal of Organic Chemistry, 65, 8264–8268. 156 Eyrisch, O. and Fessner, W.-D. (1995) Angewandte Chemie-International Edition in English, 34, 1639–1641. 157 Schultz, M., Waldmann, H., Kunz, H. and Vogt, W. (1990) Liebigs Annalen, 1019–1024. 158 Myles, D.C., Andrulis, P.J.I. and Whitesides, G.M. (1991) Tetrahedron Letters, 32, 4835–4838. 159 Chenevert, R. and Dasser, M. (2000) Journal of Organic Chemistry, 65, 4529–4531. 160 Chenevert, R., Lavoie, M. and Dasser, M. (1997) Canadian Journal of Chemistry, 75, 68–73. 161 Matsumoto, K., Shimagaki, M., Nakata, T. and Oishi, T. (1993) Tetrahedron Letters, 34, 4935–4938. 162 Shimagaki, M., Muneshima, H., Kubota, M. and Oishi, T. (1993) Chemical and Pharmaceutical Bulletin, 41, 282–286. 163 Phung, A.N., Zannetti, M.T., Whited, G. and Fessner, W.-D. (2003) Angewandte Chemie International Edition, 42, 4821–4824. 164 Sch€orken, U. and Sprenger, G.A. (1998) Biochimica et Biophysica Acta, 1385, 229–243.

References 165 Turner, N.J. (2000) Current Opinion in Biotechnology, 11, 527–531. 166 Pohl, M., Lingen, B. and M€ uller, M. (2002) Chemistry- A European Journal, 8, 5288–5295. 167 Morris, K.G., Smith, M.E.B., Turner, N.J., Lilly, M.D., Mitra, R.K. and Woodley, J.M. (1996) Tetrahedron: Asymmetry, 7, 2185–2188. 168 Effenberger, F., Null, V. and Ziegler, T. (1992) Tetrahedron Letters, 33, 5157–5160. 169 Kobori, Y., Myles, D.C. and Whitesides, G.M. (1992) Journal of Organic Chemistry, 57, 5899–5907. 170 Bolte, J., Demuynck, C. and Samaki, H. (1987) Tetrahedron Letters, 28, 5525–5528. 171 Zimmermann, F.T., Schneider, A., Sch€ orken, U., Sprenger, G.A. and Fessner, W.-D. (1999) Tetrahedron: Asymmetry, 10, 1643–1646. 172 Shaeri, J., Wohlgemuth, R. and Woodley, J.M. (2006) Organic Process Research and Development, 10, 605–610. 173 Taylor, S.V., Vu, L.D., Begley, T.P., Sch€ orken, U., Grolle, S., Sprenger, G.A., Bringer-Meyer, S. and Sahm, H. (1998) Journal of Organic Chemistry, 63, 2375–2377. 174 Humphrey, A.J., Parsons, S.F., Smith, M.E.B. and Turner, N.J. (2000) Tetrahedron Letters, 41, 4481–4485. 175 Gonzalez-Garcia, E., Helaine, V., Klein, G., Schuermann, M., Sprenger, G.A., Fessner, W.-D. and Reymond, J.-L. (2003) Chemistry- A European Journal, 9, 893–899. 176 Rogers, P.L., Shin, H.S. and Wang, B. (1997) Advances in Biochemical Engineering/Biotechnology, 56, 33–59. 177 Iding, H., D€ unnwald, T., Greiner, L., Liese, A., M€ uller, M., Siegert, P., Gr€ otzinger, J., Demir, A.S. and Pohl, M. (2000) Chemistry- A European Journal, 6, 1483–1495. 178 Demir, A.S., Sesenoglu, Ö., Eren, E., Hosrik, B., Pohl, M., Janzen, E., Kolter, D., Feldmann, R., D€ unkelmann, P. and M€ uller, M. (2002) Advanced Synthesis and Catalysis, 344, 96–103.

179 Barbas, C.F., Wang, Y.F. and Wong, C.-H. (1990) Journal of the American Chemical Society, 112, 2013–2014. 180 Chen, L., Dumas, D.P. and Wong, C.-H. (1992) Journal of the American Chemical Society, 114, 741–748. 181 Gijsen, H.J.M. and Wong, C.-H. (1995) Journal of the American Chemical Society, 117, 2947–2948. 182 Liu, J., DeSantis, G. and Wong, C.-H. (2002) Canadian Journal of Chemistry, 80, 643–645. 183 DeSantis, G., Liu, J., Clark, D.P., Heine, A., Wilson, I.A. and Wong, C.-H. (2003) Bioorganic and Medicinal Chemistry, 11, 43–52. 184 Tischer, W., Ihlenfeldt, H.-G., Barzu, O., Sakamoto, H., Pistotnik, E., Marliere, P. and Pochet, S. (2001) WO 0114566; Chemical Abstracts, 134, 208062. 185 Ouwerkerk, N., Van Boom, J.H., Lugtenburg, J. and Raap, J. (2000) European Journal of Organic Chemistry, 861–866. 186 Liu, J. and Wong, C.-H. (2002) Angewandte Chemie International Edition, 41, 1404–1407. 187 Gijsen, H.J.M. and Wong, C.-H. (1995) Journal of the American Chemical Society, 117, 7585–7591. 188 Wong, C.-H., Garcia-Junceda, E., Chen, L.R., Blanco, O., Gijsen, H.J.M. and Steensma, D.H. (1995) Journal of the AmericanChemicalSociety,117,3333–3339. 189 Liu, J., Hsu, C.-C. and Wong, C.-H. (2004) Tetrahedron Letters, 45, 2439–2441. 190 Greenberg, W.A., Varvak, A., Hanson, S.R., Wong, K., Huang, H., Chen, P. and Burk, M.J. (2004) Proceedings of the National Academy of Sciences of the United States of America, 101, 5788–5793. 191 Jennewein, S., Sch€ urmann, M., Wolberg, M., Hilker, I., Luiten, R., Wubbolts, M. and Mink, D. (2006) Journal of Biotechnology, 1, 537–548. 192 Liu, J.Q., Dairi, T., Itoh, N., Kataoka, M., Shimizu, S. and Yamada, H. (2000) Journal of Molecular Catalysis B: Enzymatic, 10, 107–115.

j317

j 10 Aldolases: Enzymes for Making and Breaking C----C Bonds

318

193 Kimura, T., Vassilev, V.P., Shen, G.J. and Wong, C.-H. (1997) Journal of the American Chemical Society, 119, 11734–11742. 194 Saeed, A. and Young, D.W. (1992) Tetrahedron, 48, 2507–2514. 195 Lotz, B.T., Gasparski, C.M., Peterson, K. and Miller, M.J. (1990) Journal of the Chemical Society, Chemical Communications, 16, 1107–1109. 196 Steinreiber, J., Fesko, K., Reisinger, C., Sch€ urmann, M., van Assema, F., Wolberg, M., Mink, D. and Griengl, H. (2007) Tetrahedron, 63, 918–926. 197 Vassilev, V.P., Uchiyama, T., Kajimoto, T. and Wong, C.-H. (1995) Tetrahedron Letters, 36, 4081–4084. 198 Ura, D., Hashimukai, T., Matsumoto, T. and Fukuhara, N. (1991) EP 421477; Chemical Abstracts, 115, 90657. 199 Anderson, D.M. and Hsiao, H.-H. (1992) In Biocatalytic Production of Amino Acids and Derivatives (eds D. Rozzell and F. Wagner), Hanser, Munich, pp. 23–41. 200 Bycroft, M., Herbert, R.B. and Ellames, G.J. (1996) Journal of the Chemical Society, Perkin Transactions 1, 2439–2442. 201 Fujii, M., Miura, T., Kajimoto, T. and Ida, Y. (2000) Synlett 7, 1046–1048. 202 Miura, T., Fujii, M., Shingu, K., Koshimizu, I., Naganoma, J., Kajimoto, T. and Ida, Y. (1998) Tetrahedron Letters, 39, 7313–7316. 203 Shibata, K., Shingu, K., Vassilev, V.P., Nishide, K., Fujita, T., Node, M.,

204

205

206

207

208

209

210 211

212

213

Kajimoto, T. and Wong, C.-H. (1996) Tetrahedron Letters, 37, 2791–2794. Nishide, K., Shibata, K., Fujita, T., Kajimoto, T., Wong, C.-H. and Node, M. (2000) Heterocycles, 52, 1191–1201. Wu, S.H., Shimazaki, M., Lin, C.C., Qiao, L., Moree, W.J., Weitz-Schmidt, G. and Wong, C.-H. (1996) Angewandte ChemieInternational Edition in English, 35, 88–90. Steinreiber, J., Sch€ urmann, M., Wolberg, M., van Assema, F., Reisinger, C., Fesko, K., Mink, D. and Griengl, H. (2007) Angewandte Chemie International Edition, 46, 1624–1626. Herbert, R.B., Wilkinson, B. and Ellames, G.J. (1994) Canadian Journal of Chemistry, 72, 114–117. Liu, J.Q., Odani, M., Dairi, T., Itoh, N., Shimizu, S. and Yamada, H. (1999) Applied Microbiology and Biotechnology, 51, 586–591. Liu, J.Q., Odani, M., Yasuoka, T., Dairi, T., Itoh, N., Kataoka, M., Shimizu, S. and Yamada, H. (2000) Applied Microbiology and Biotechnology, 54, 44–51. Kazmeier, U. (2005) Angewandte Chemie International Edition, 44, 2186–2188. Williams, G.J., Nelson, A.S. and Berry, A. (2004) Cellular and Molecular Life Sciences, 61, 3034–3046. Lorenz, P., Liebeton, K., Niehaus, F. and Eck, J. (2002) Current Opinion in Biotechnology, 13, 572–577. Streit, W.R., Daniel, R. and Jaeger, K.-E. (2004) Current Opinion in Biotechnology, 15, 285–290.

j319

Index a acetaldehyde-dependent aldolase 276 Acetobacterium woodii 196 acetone dehydration 206 acetophenone derivatives 213 N-acetylaspartic semialdehyde 291 N-acetyl-D-mannosamine 278 N-acetylneuraminic acid aldolase 278 N-acetyl-phenylalanine methyl ester 16 acid-catalyzed DKR 102 – approach 90, 115, 135, 179 acid-catalyzed racemization 101, 109 acidic zeolite catalyst 102 acrylic moiety 291 acylated racemic alcohol 182 acylating agents 180 acylation reactions 150 acyl donor 10 Agrobacterium radiobacter 42 Alcaligenes xylosoxidans 15, 309 alcohol dehydrogenase (ADH) 235 alcohols – deracemization 124 – enzymatic transformations 150 – regioselective oxidation 231 aldehydes – aromatic 308 – chiral phosphorylated 46 – racemization 105 – reduction 216 aldol reaction 276 aliphatic open-chain monooxygenase (AOCMO) 250 alkane hydroxylases 242 alkyl aryl ketones 201 alkylation–reduction reaction 223 alkyl methoxy acetates 180 Amaryllidaceae alkaloids 262

AmberliteTM XAD 211 amines – deracemization 237 – oxidations 231 amino acids – conjugates 280 – randomization 35 – synthesis 146 aminolysis reactions – mechanism 172 aminotransferases 45 amphiphilic polymer polyethylene glycol (PEG) 9 Amycolapsis orientalis 255 anaerobic interface bioreactor 211 anion-exchange resin 103 anti-inﬂammatory ibuprofen 178 antimicrobial genes 79 antipodal lactones 247 antiviral carbocyclic nucleosides 148 aqueous-organic two-phase reaction 209 aromatic acyl donors 151 aromatic amino acid ethyl esters – kinetic resolution 6 aromatic sulfonamide 291 aryl dioxygenases 257 arylglycidic acids 145 Ascophyllum nodosum 253 Aspergillus oryzae 10 azasugar phosphonic acids 297

b Bacillus megaterium 55, 204, 240 Bacillus stearothermophilus 15, 124, 194 Baeyer–Villiger oxidations 243 Baeyer–Villiger process 48, 245 baker’s yeast 15, 201 base-catalyzed transhydrocyanation 103

Asymmetric Organic Synthesis with Enzymes. Edited by Vicente Gotor, Ignacio Alfonso, and Eduardo Garca-Urdiales Copyright Ó 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 978-3-527-31825-4

j Index

320

bay structural motifs, see fjord structural motifs Beauveria bassiana 238 benzoylformate decarboxylase 304 bicyclic carbohydrate structures 293 biocatalytic method 176, 247 biocatalytic reactions 3 biocatalytic reduction system 193, 199 biodiversity mining approach 65 biohalogenation metabolites 263 bioisosteric methylene phosphonate analog 295 bioisosteric phosphonate nucleophile 297 biooxidized benzoic acid 262 biotechnological processes – complete hydrolysis 77 biphasic water–tetradecane solvent system 196 Bischler–Napieralski-type cyclization reaction 261–262 Bromeliad tank 79 2-bromo-4-ﬂuoroacetophenone 202 bromo-iodobenzene 260 Bronsted–Lowry acids 16 tert-butyl disulﬁde 256 tert-butyl 4-methyl-3,5-dioxohexanoate 222 tert-butyl methyl ether (TBME) 184

c Cahn–Ingold–Prelog system 96 calcium hydroxide 152 Caldariomyces fumago 263 Candida antarctica lipase A (CALA) 181 Candida antarctica lipase B (CALB) 10, 13, 172 Candida magnoliae 200 Candida parapsilosis carbonyl reductase (CPCR) 203, 219 Candida rugosa 15 carboxylic acid 39, 126 – enzymatic aminolysis 171, 174 – esteriﬁcation 140 catalytic antibodies 66–67 catalytic promiscuity – concept 69 chemical macrolactonizations 142 chemoenzymatic epoxidation reactions 243 chemoenzymatic oxygenation sequence 262 chiral amines 119, 181 – preparation 171 chiral aziridines 256 chiral diendiols 241 chiral hydroxy-aldehyde 233 chiral hydroxyl-ketones 233 chiral lactone intermediates 243

chiral lactone product 248 chiral ligands 89 chiral nonracemic epoxides 157 chiral organic compounds 21 chiral phosphates 45 chlorobenzene dioxygenase (CDO) 258 4-chloro-3-oxobutanoate 205 p-chlorophenyl acetate 94 cholesterol gallstone-dissolving agent 211 chrysanthemic acids 145 closed-loop systems 230 combinatorial active-site saturation test (CAST) 25, 35 combinatorial multiple-cassette mutagenesis (CMCM) 30 L-conﬁgured aldohexoses 295 Corallina ofﬁcinalis 253 Corallina pilulifera 253 Cunninghamella echinulata 236 Curtius-type reaction 138 cyclic oxidation–reduction system 117 cyclic structural motifs 237 cyclobutanone structural motif 251 cyclododecanone monooxygenase (CDMO) 244 cyclohexanone monooxygenases (CHMO) 48 cyclooleﬁnic precursor 299 cyclopentadecanone monooxygenase (CPDMO) 244 cyclopentanone monooxygenase (CPMO) 48, 52 cytochrome P450 enzymes 55

d cis-decaline system 245 Dendroctonus brevicomis 300 3-deoxy-D-arabino-heptulosonic acid 7phosphate (DAHP) 291 deracemization process 115–116, 122, 223 detection technologies 73 DHAP nucleophile 294 diastereomeric switch sequences 260 diastereomer pseudoephedrine 304 diastereoselective dynamic kinetic resolution process 309 diastereoselective L-tyrosine decarboxylase 309 Diels–Alder reaction 96, 97, 156 diethylmaleate 177 dihexyl ether 211 direct enzymatic synthesis 175 DNA ampliﬁcation method 23 DNA shufﬂing 27, 30, 31, 42, 46, 56

Index dried-cell system 194, 216 dynamic combinatorial library 280

e electrochemical cofactor recycling 230 electron-rich aromatic cores 263 electrophilic aldehyde acceptor 276 electrophilic halogenating species 264 enantioconvergent process 127, 130, 159 enantioconvergent transformations 115, 137 enantiomerically pure compounds – synthesis 89 enantiopure amines 180 enantiopure ketohexose 1-phosphates 290 enantioselective alkyl sulfatases 129 enantioselective amidation process 174 enantioselective antibodies 58 enantioselective enzymatic desymmetrization 135 enantioselective hydrogenation catalyst 70 Enterobacter agglomerans 233 enzymatic aldolization technique 294 enzymatic alkoxycarbonylation 185 enzymatic aminolysis reaction 176, 177 enzymatic epoxidation 241 enzymatic hydrolysis, see transesteriﬁcation process enzymatic oxygenation process 253 enzyme-bound ﬂavin (FAD) 48 enzyme-bound pyridoxamine phosphate intermediate 45 enzyme-catalyzed – acylation 177 – deracemization processes 130 – idol reactions 290 – oxidation 108 – racemization 106 – stereoselective hydrolysis 141 – synthetic transformations 91 enzyme hydration 8 enzyme lyophilization 9 enzyme-mediated – chiral sulfoxidation 253 – halogenation 263 – oxidation reactions 229 enzyme membrane reactor (EMR) 279 enzyme promiscuity 56 enzyme selectivity – modulation 14 enzyme–substrate interaction 9 enzyme–substrate transition state complexes 3 enzyme-triggered enantioconvergent cascade reaction 160

epileptic seizures 262 epoxide biohydrolyses 158 epoxide hydrolases (EH) 41, 126, 157 – use 128 error-prone polymerase chain reaction (epPCR) 23 Escherichia coli 21, 73, 120, 244, 283 esters – aminolysis 176 – enzymatic resolution 178 – alcoholysis 140 – hydrolysis 137 ethyl acetate 180 ethyl 4-chloro-3-oxobutanoate (COBE) 200, 203 ethyl 2-methyl-3-oxobutanoate 200 ethyl 2-oxo-4-phenylbutyrate 201 N-ethyl-pyridinium triﬂuoroacetate 15

f family shufﬂing 27 fatty acid esters – hydrolysis 35 Fe-heme-containing enzymes 55 ﬁlter-press electrochemical reactor 198 fjord structural motifs 258 ﬂavin monooxygenases 263 ﬂavoprotein reductase 257 ﬂuorescence activating cell sorting (FACS) 74 ﬂuorous phase separation techniques 135 formate dehydrogenase (FDH) 195 fossil fuel-derived polymers 79 four-substituted cyclohexanone derivatives 54 fructose 1,6-bisphosphate aldolase 204, 285 FucA catalyzed aldol reactions 295 L-fucose analogs 295 L-fucose isomerases 294 furanose isomers, see pyranose isomers

g D-galacto-conﬁgured

carbohydrates 286 galactose oxidase (GAOX) 233, 293 Gd-labeled cholanoic acids 232 genetic algorithms, see neural networks genetic methods 201 genomic fusion protein libraries 201 Geotrichum candidum 194 Gluconobacter oxydans 234 glucose dehydrogenase (GDH) gene 204, 232 glutathione S-transferase (GST)-fusion proteins 201 D-glyceraldehyde acetonide 308

j321

j Index

322

D-glyceraldehyde

3-phosphate (GA3P) 282, 285, 305 glycerol hydratases 79 L-glycerol 3-phosphate 289 glycerol phosphate oxidases (GPO) 289 glycine aldolases 308 glycine-dependent enzymes 276 glycine-utilizing aldolases 277 glyco-processing enzymes 296 C-glycosidic aldehydes 292 O-glycosidic aldehydes, see C-glycosidic aldehydes goniofufurone analogs 245 green chemistry 229

h habitat-related halotolerant characteristics 76 habitat-related properties 77 halogenated natural products 263 halogenation reactions 263 heteroatom oxidations 253 hetero-bicyclic substrate 245 heterologous gene expression 75 high-throughput colorimetric screen 120 high-throughput screening methods 21, 27, 109 horse liver alcohol dehydrogenase (HLADH) 199, 233 cis-hydrindane substrate 245 hydrogen transfer catalyst 95 hydrolase-catalyzed reactions 134 hydrolytic enzyme cutinase 12 hydrolytic enzymes 188 hydrolytic reactions 150 hydrophobic bases 9 hydrophobic polymer XAD-7 212 hydrophobic products 209 hydrophobic resin – use 211 hydrophobic sol–gel materials 9 hydrophobic substrate 209 hydroxyacetone monooxygenase (HAPMO) 254 4-hydroxycyclohexanone 248 hydroxysteroid dehydrogenases (HSHDs) 231 hyperthermophilic esterase 39 hyperthermophilic sulfur-metabolizers 129

i iminocyclitol phosphonic acids 297 immobilization methods 134 immobilized lipases 9 immunosuppressive lipid mycestericind 308

indole alkaloids alloyohimbane 247 indolizidine-type glycosidase inhibitor castanospermine 281 insect pheromones 249 in situ isomerization 279 in situ racemization 146 in vitro coupled transcription/translation 38 ionic liquids 14–15, 215 – use 174 isobutyl compound – hydrolysis 159 iterative saturation mutagenesis (ISM) 25

k Kazlauskas’ rule 96, 180, 181, 183 2-keto-3-deoxy aldonic acid intermediates 278 2-keto-3-deoxy-manno-octosonate 278 2-keto-3-deoxy-6-phospho-D-galactonate 284 2-keto-3-deoxy-6-phospho-D-gluconate 282 keto methyl ester 203 keto-reductase-screening kit 201 ketose monosaccharides 294 kinetic resolution 90, 134, 171 – drawbacks 135 – efﬁcacy 135 Klebsiella aerogenes 308 Klebsiella pneumoniae IFO 3319 200 Klibanov’s group 9, 10 Kluyveromyces lactis 203 Kluyveromyces marxianus 221 Koshland’s model of induced ﬁt 33

l Lactobacillus brevis alcohol dehydrogenase (LBADH) 219 Lactobacillus keﬁr 235 leucine methyl group 264 lichen macrolactone 301 lipase-catalyzed – acylation 93, 181 – ammonolysis 105 – esteriﬁcation 141 – hydrolysis 93 – methanolysis 140 – reaction 28, 172 lipid–water interface 133 lipolytic biocatalysts 77 lipolytic genes 77 long-chain N-acyltyrosine antibiotics 79

m macrolide antibiotic pentamycin 301 magnetic resonance imaging 232

Index Meerwein–Ponndorf–Verley–Oppenauer (MPVO) reaction 94, 96 membrane electrochemical reactors 198 meso cyclohexene diester 5 meso-epoxides 157, 162 – asymmetric hydrolysis 157 metabolic pathways 45, 257 metagenome-derived DNA libraries 72 metagenome-derived enzymes 75 metagenomic genes 75, 80 – expression 75 – library 75, 77, 78, 79 metal-assisted lipase-mediated strategies 235 metal-catalyzed racemization 92 methylene phosphonate 289 methyl 7-ketolithocholate (Me-7KLCA) 211 methyl vinyl ketone 208 mevinic acid lactone 306 Michaelis–Menten constant 209 Michaelis–Menten kinetics 90 Microbacterium campoquemadoensis 200, 201 monoamine oxidases 54, 119, 120 Mortierella isabellina 253 Mucor miehei lipase (MML) 153 mutagenesis method 21, 23, 28–29, 30 – application 31 – biological methods 23 – saturation 31, 41 Mycobacterium tuberculosis 244, 252

n NADH-dependent ketone reductase 195 naphthalene dioxygenase (NDO) 254, 301 neural networks 56 neuraminidase inhibitors 280 nicotinamide adenine dinucleotide phosphate (NADPH) 229 nikkomycin 283 nitrile amidase system 144 nitrile hydratase, see nitrile amidase system nitriles – biocatalytic hydrolysis 40, 144 – chemical hydrolysis 78 p-nitrophenyl glycidyl ether 42 nonchiral amides – preparation 176 nicotinamide systems 230 – nonphosphorylated 230 – phosphorylated 230 nucleophilic serine residue 69 NusA protein 68

o one-pot double aminolysis 183 one-pot resolution 182 one-pot sequential bienzymatic strategy 159 one-pot strategy 235 optimized fermenter technology 241 organic-aqueous two-phase system 204 organic media 5, 7 organic solvent systems 7, 13 – enzyme properties 8 organocatalysts 21 organometallic complexes 92 organometallic substrates 155 – resolution 152 orthoformates 141 ortho-substituted acetophenones 211 oxidation–reduction deracemization process 118 oxidoreductases 79 oxirane-containing nitriles – enantioselective transformations 144

p palmitoylputrescine 79 para-substituted acetophenones 211 para-substituted analogs 123 para-substituted a -methylstyrene oxides – hydrolysis 159 para-substituted phenylethanol derivatives 223 parental genes – self-hybridization 27 PCR-ampliﬁcation 27 Pd-catalyzed hydrogenation 297 Pd-catalyzed racemization 98 pentose catabolism, see hexose catabolism pentose monophosphates, see hexose monophosphates peptide deformylases – natural role 147 phenylacetaldehyde reductase (PAR) 216 phenylacetone monooxygenase (PAMO) 243 phenylcyclohexanone monooxygenase (PCHMO) 51 4-phenyl-2-oxo-but-3-enoates 220 phenylpropanoid acid 234 phenylserine derivative 309 phosphate buffer system 124 phosphonic acid esters 45 phosphorylated nicotinamide systems 230 phosphotriesterases 45 photochemical methods 196, 247 Pichia methanolica SC 13825 203 pig liver esterase (PLE) 137

j323

j Index

324

pilot-scale bioepoxidations 241 pilot-scale EMR 279 plant cell culture 223 podophyllotoxin 156 polar hydrophilic solvents 134 poly-ethylene glycol-acrylamide (PEGA) 175 polyketide-derived natural products 156 polymerase chain reaction (PCR) 23 polymer-supported derivative 96 polyoxygenated cycloalkanes 260 polypropionate chains 155 porcine kidney D-AAO 237 porcine pancreas lipase (PPL) 143 primary protein sequence 247 prochiral chromanedimethanol derivative – stereoselective acylation 155 prochiral diester malonates 16 prochiral ketones 220 prochiral model substrates 195 – desymmetrization 50 prokaryotic diversity 71 proteins 9, 13, 71, 75, 264, 265 – noncatalytic 70 protein engineering 33 protein–protein stacking 9 protein sequence space 24, 31 Proteus myxofaciens 237 proximal lactones 252 Pseudomonas aeruginosa lipase (PAL) 28 Pseudomonas cepacia lipase (PSL) 10, 94, 142, 178 Pseudomonas diminuta 45 Pseudomonas ﬂuorescens lipase (PFL) 94 Pseudomonas frederickberensis 254 Pseudomonas oleovorans 242 Pseudomonas putida 239, 301 pyranose isomers 290 pyranose oxidases (P2Os) 232 pyridine system 239 pyridoxal phosphate–dependent enzymes 308 pyridyl oxiranes 158 Pyrococcus furiosus 195 pyrrolidinol derivatives – synthesis 179 pyruvate aldolases 278, 281 pyruvate decarboxylase (PDC) 303

q quaternary ammonium group 67

r rabbit muscle aldolase (RAMA) 285 racemic azido-substituted cyclohexenediol precursor 299 racemic epoxides

– deracemization 126 racemic esters 38 racemic 2-hydroxyaldehydes 293, 302 racemic ketones 50 racemic b-lactone 142 racemic mandelic acid derivatives – deracemization 125 racemic-2-methylcyclohexanone 220 racemic methyl decanoic acids 141 racemic substrate – enzymatic resolution 3 – kinetic resolutions 248 racemic sulfates – enantioconvergent hydrolysis 129 racemize secondary alcohols 96 recombinant cytochrome P450 monooxygenases – hydroxylations 239 recombinant expression system 243, 254 redox biocatalysis 229 regiodivergent biooxidations 251 reversible Diels–Alder cyclization, see Mitsunobu inversion reversible glycerol phosphorylation 289 L-rhamnulose1-phosphate aldolase 286 Rh-catalyzed hydrogenation 89 Rhizomucor miehei lipase (Lipozyme) 100 Rhodococcus erythropolis 195, 234 Rhodococcus ruber DSM 44541 220, 234 RhuA catalyzed aldol reactions, see FucA catalyzed aldol reactions ring-closing metatheses (RCM) 154 ruthenium-based organometallic catalysts 135 ruthenium-catalyzed oxidation 96 ruthenium-catalyzed racemization 97 ruthenium complex 93, 196 ruthenium precatalyst 95

s Saccharomyces cerevisiae 197, 244 schiff-base intermediates 104 screening techniques 200 sec-alkyl sulfate esters – enantioselective hydrolysis 152 secondary alcohols – deracemization 235 – microbial deracemization 122, 124 – subtilisin-catalyzed resolution 13 secondary amines – kinetic resolution 185 selective inhibitor 207 selenium oxidation 256 sequential aldolization–hydrogenation 306

Index serine hydrolase mechanism 172 serine hydroxymethyltransferase (SHMT) 308 Shingomonas sp 242 Shvo’s ruthenium complex 94 SIGEX system 74 sodium borohydride 118 soil metagenomic library 79 – functional screening 75 soluble organic solvent 209 solvent–enzyme complexes 13 Sphingomonas yanoikuyae 223, 236, 257 spiro-annulated sugars 290 Sporobolomyces salmonicolor 216 Staphylococcus carnosus 285 stereo-divergent chemoenzymatic syntheses 296 stereoinversion mechanism 115 stereoselective dehydrogenase 231 stereoselective enzymatic transformation 156 steroid ABCD-ring system 238 stoichiometric oxidant 96 Strecker reactions 38, 145 Streptomyces amakusaensis 310 structure–function relationship 80 styrene oxides – enantioconvergent hydrolysis 128 a-substituted acetaldehyde 306 b-substituted cycloketones 252 6-substituted D-fructofuranoside derivatives 291 4-substituted-2-thiopropanoic acids – deracemization 126 substrate feeding–product removal (SFPR) concept 244 sugar phosphonates 289 sulfate esters – hydrolysis 152 sulfoxidation reactions 245 surface-modiﬁed electrodes 198 symmetrical dienophiles 262 synthetic enzyme models 71 synthetic libraries 65 p-system geometry 177

t D-tagatose 1,6-bisphosphate aldolase 286 tertiary alkoxy intermediate 93 tetrachloro-benzene dioxygenase 262 tetrahydrofuran natural products 245 – goniofufurone 245 – kumausyne 245 Thermoanaerobacter ethanolicus alcohol dehydrogenase 208

Thermobiﬁda fusca 51, 254 Thermus thermophilus 240 thiamine diphosphate-dependent enzymes 303, 304 three-point binding model 281 toluene dioxygenase 254 transesteriﬁcation process 92, 182 transketolase assays 292 transketolase enzymes 302 triose phosphate isomerase 288 tropane alkaloid ferruginine – divergent synthesis 139 Trypanosoma brucei inhibitor 291 TSA enzyme inhibitors 67 TSA mimicry approach 69 two-step lipase-catalyzed reactions 187

u UV–vis–based screening system 28, 29

v vanadium-dependent haloperoxidases 264 vicinal D-threo diols 302 vinyl acetate 150 viologen-diaphorase 198 volume efﬁcient process 175

w water-absorbing polymer 210 water-in-oil emulsion systems 287 water-miscible organic cosolvents 5, 7 water-soluble monoester monoacid 135 water-stable sodium cyanoborohydride 118 white-rot fungus 223 whole-cell-mediated biotransformations 253, 257, 262 whole-cell system 39 WT hydantoinases 39

x X-ray crystal structure 41 X-ray diffraction 239, 243 X-ray structure 35 xylene monooxygenases 242 D-xylulose 5-phosphate 303

y Yarrowia lipolytica 124

z Zanamivir 279 zinc tetrakis(4-methylpyridyl) porphyrin (ZnTMPyP) 197

j325